58 Commits
v1.1 ... v2.0

Author SHA1 Message Date
6f5afbc6ec Update readme with CLI arguments 2022-02-27 17:01:12 -06:00
fb4d22e7a4 Update readme with CLI arguments 2022-02-27 17:00:54 -06:00
e10350c214 Update readme with CLI arguments 2022-02-27 16:56:58 -06:00
b1155f8100 Format -help CLI option 2022-02-27 16:53:46 -06:00
12b003a69f Add -help CLI option 2022-02-27 16:45:30 -06:00
32c5bcaaff Deactivate file I/O announcement for CLI 2022-02-27 16:16:24 -06:00
2485ac4cf6 Add getters to MatchingResult 2022-02-27 16:15:26 -06:00
05556bce0c Add units to metadata 2022-02-27 16:08:59 -06:00
a822f69ea4 Control verbose output 2022-02-27 16:07:17 -06:00
3d1f8668ee Control verbose output 2022-02-27 16:03:57 -06:00
40c743308b Initialize wells 2022-02-27 15:54:47 -06:00
5246cc4a0c Re-implement command line options 2022-02-27 15:35:07 -06:00
a5f7c0641d Refactor for better encapsulation with CellSamples 2022-02-27 14:51:53 -06:00
8ebfc1469f Refactor plate to fill its own wells in its constructor 2022-02-27 14:25:53 -06:00
b53f5f1cc0 Refactor plate to fill its own wells in its constructor 2022-02-27 14:17:16 -06:00
974d2d650c Refactor plate to fill its own wells in its constructor 2022-02-27 14:17:11 -06:00
6b5837e6ce Add Vose's alias method to to-dos 2022-02-27 11:46:11 -06:00
b4cc240048 Update Readme 2022-02-26 11:03:31 -06:00
ff72c9b359 Update Readme 2022-02-26 11:02:23 -06:00
88eb8aca50 Update Readme 2022-02-26 11:01:44 -06:00
98bf452891 Update Readme 2022-02-26 11:01:20 -06:00
c2db4f87c1 Update Readme 2022-02-26 11:00:18 -06:00
8935407ade Get rid of GraphML reader, those files are larger than serialized files 2022-02-26 10:38:10 -06:00
9fcc20343d Fix GraphML writer 2022-02-26 10:36:00 -06:00
e4d094d796 Adding GraphML output to options menu 2022-02-24 17:22:07 -06:00
f385ebc31f Update vertex class 2022-02-24 16:25:01 -06:00
8745550e11 add MWM algorithm type to matching metadata 2022-02-24 16:24:48 -06:00
41805135b3 remove unused import 2022-02-24 16:04:30 -06:00
373a5e02f9 Refactor to make CellSample class more self-contained 2022-02-24 16:03:49 -06:00
7f18311054 fix typos 2022-02-24 15:55:32 -06:00
bcb816c3e6 Reformat TODO 2022-02-24 15:48:10 -06:00
dad0fd35fd Update readme to reflect wells with random population implemented 2022-02-24 15:47:08 -06:00
35d580cfcf Update readme to reflect wells with random population implemented 2022-02-24 15:45:03 -06:00
ab8d98ed81 Update readme to reflect new default caching behavior. 2022-02-24 15:39:15 -06:00
3d9890e16a Change GraphModificationFunctions to only save edges if graph data is cached 2022-02-24 15:32:27 -06:00
dd64ac2731 Change GraphModificationFunctions to interface 2022-02-24 15:18:09 -06:00
a5238624f1 Change default graph caching behavior to false 2022-02-24 15:14:28 -06:00
d8ba42b801 Fix Algorithm Options menu output 2022-02-24 14:59:08 -06:00
8edd89d784 Added heap type selection, fixed error handling 2022-02-24 14:48:19 -06:00
2829b88689 Update readme to reflect caching changes 2022-02-24 12:47:26 -06:00
108b0ec13f Improve options menu wording 2022-02-24 12:42:09 -06:00
a8b58d3f79 Output new setting when changing options 2022-02-24 12:38:15 -06:00
bf64d57731 implement option menu for file caching 2022-02-24 12:30:47 -06:00
c068c3db3c implement option menu for file caching 2022-02-23 20:35:31 -06:00
4bcda9b66c update readme 2022-02-23 13:22:04 -06:00
17ae763c6c Generate populations correctly 2022-02-23 10:37:40 -06:00
decdb147a9 Cache everything 2022-02-23 10:30:42 -06:00
74ffbfd8ac make everything use same random number generator 2022-02-23 09:29:21 -06:00
08699ce8ce Change output order to match interactive UI 2022-02-23 08:56:09 -06:00
69b0cc535c Error checking 2022-02-23 08:55:07 -06:00
e58f7b0a55 checking for possible divide by zero error. 2022-02-23 08:54:14 -06:00
dd2164c250 implement sample plates with random well populations 2022-02-23 08:14:17 -06:00
7323093bdc change "getRandomNumber" to "getRandomInt" for consistency. 2022-02-23 08:13:52 -06:00
f904cf6672 add more data caching code 2022-02-23 08:13:06 -06:00
3ccee9891b change "concentrations" to "populations" for consistency 2022-02-23 08:12:48 -06:00
40c2be1cfb create populations string correctly 2022-02-23 08:11:01 -06:00
4b597c4e5e remove old testing code 2022-02-23 08:10:35 -06:00
b2398531a3 Update readme 2022-02-23 05:11:36 +00:00
19 changed files with 1126 additions and 561 deletions

129
readme.md
View File

@@ -12,7 +12,7 @@ Unlike pairSEQ, which calculates p-values for every TCR alpha/beta overlap and c
against a null distribution, BiGpairSEQ does not do any statistical calculations
directly.
BiGpairSEQ creates a [simple bipartite weighted graph](https://en.wikipedia.org/wiki/Bipartite_graph) representing the sample plate.
BiGpairSEQ creates a [weighted bipartite graph](https://en.wikipedia.org/wiki/Bipartite_graph) representing the sample plate.
The distinct TCRA and TCRB sequences form the two sets of vertices. Every TCRA/TCRB pair that share a well
are connected by an edge, with the edge weight set to the number of wells in which both sequences appear.
(Sequences present in *all* wells are filtered out prior to creating the graph, as there is no signal in their occupancy pattern.)
@@ -29,17 +29,13 @@ Unfortunately, it's a fairly new algorithm, and not yet implemented by the graph
So this program instead uses the Fibonacci heap-based algorithm of Fredman and Tarjan (1987), which has a worst-case
runtime of **O(n (n log(n) + m))**. The algorithm is implemented as described in Melhorn and Näher (1999).
The current version of the program uses a pairing heap instead of a Fibonacci heap for its priority queue,
which has lower theoretical efficiency but also lower complexity overhead, and is often equivalently performant
in practice.
## USAGE
### RUNNING THE PROGRAM
[Download the current version of BiGpairSEQ_Sim.](https://gitea.ejsf.synology.me/efischer/BiGpairSEQ/releases)
BiGpairSEQ_Sim is an executable .jar file. Requires Java 11 or higher. [OpenJDK 17](https://jdk.java.net/17/)
BiGpairSEQ_Sim is an executable .jar file. Requires Java 14 or higher. [OpenJDK 17](https://jdk.java.net/17/)
recommended.
Run with the command:
@@ -52,34 +48,70 @@ For example, to run the program with 32 gigabytes of memory, use the command:
`java -Xmx32G -jar BiGpairSEQ_Sim.jar`
Once running, BiGpairSEQ_Sim has an interactive, menu-driven CLI for generating files and simulating TCR pairing. The
main menu looks like this:
There are a number of command line options, to allow the program to be used in shell scripts. For a full list,
use the -help flag:
`java -jar BiGpairSEQ_Sim.jar -help`
If no command line arguments are given, BiGpairSEQ_Sim will launch with an interactive, menu-driven CLI for
generating files and simulating TCR pairing. The main menu looks like this:
```
--------BiGPairSEQ SIMULATOR--------
ALPHA/BETA T CELL RECEPTOR MATCHING
USING WEIGHTED BIPARTITE GRAPHS
USING WEIGHTED BIPARTITE GRAPHS
------------------------------------
Please select an option:
1) Generate a population of distinct cells
2) Generate a sample plate of T cells
3) Generate CDR3 alpha/beta occupancy data and overlap graph
4) Simulate bipartite graph CDR3 alpha/beta matching (BiGpairSEQ)
8) Options
9) About/Acknowledgments
0) Exit
```
### OUTPUT
By default, the Options menu looks like this:
```
--------------OPTIONS---------------
1) Turn on cell sample file caching
2) Turn on plate file caching
3) Turn on graph/data file caching
4) Turn off serialized binary graph output
5) Turn on GraphML graph output
6) Maximum weight matching algorithm options
0) Return to main menu
```
### INPUT/OUTPUT
To run the simulation, the program reads and writes 4 kinds of files:
* Cell Sample files in CSV format
* Sample Plate files in CSV format
* Graph and Data files in binary object serialization format
* Graph/Data files in binary object serialization format
* Matching Results files in CSV format
When entering filenames, it is not necessary to include the file extension (.csv or .ser). When reading or
writing files, the program will automatically add the correct extension to any filename without one.
These files are often generated in sequence. When entering filenames, it is not necessary to include the file extension
(.csv or .ser). When reading or writing files, the program will automatically add the correct extension to any filename
without one.
To save file I/O time, the most recent instance of each of these four
files either generated or read from disk can be cached in program memory. When caching is active, subsequent uses of the
same data file won't need to be read in again until another file of that type is used or generated,
or caching is turned off for that file type. The program checks whether it needs to update its cached data by comparing
filenames as entered by the user. On encountering a new filename, the program flushes its cache and reads in the new file.
(Note that cached Graph/Data files must be transformed back into their original state after a matching experiment, which
may take some time. Whether file I/O or graph transformation takes longer for graph/data files is likely to be
device-specific.)
The program's caching behavior can be controlled in the Options menu. By default, all caching is OFF.
The program can optionally output Graph/Data files in .GraphML format (.graphml) for data portability. This can be
turned on in the Options menu. By default, GraphML output is OFF.
---
#### Cell Sample Files
Cell Sample files consist of any number of distinct "T cells." Every cell contains
four sequences: Alpha CDR3, Beta CDR3, Alpha CDR1, Beta CDR1. The sequences are represented by
@@ -97,7 +129,6 @@ Comments are preceded by `#`
Structure:
---
# Sample contains 1 unique CDR1 for every 4 unique CDR3s.
| Alpha CDR3 | Beta CDR3 | Alpha CDR1 | Beta CDR1 |
|---|---|---|---|
@@ -121,15 +152,18 @@ Options when making a Sample Plate file:
* Standard deviation size
* Exponential
* Lambda value
* (Based on the slope of the graph in Figure 4C of the pairSEQ paper, the distribution of the original experiment was exponential with a lambda of approximately 0.6. (Howie, et al. 2015))
* *(Based on the slope of the graph in Figure 4C of the pairSEQ paper, the distribution of the original experiment was approximately exponential with a lambda ~0.6. (Howie, et al. 2015))*
* Total number of wells on the plate
* Number of sections on plate
* Number of T cells per well
* per section, if more than one section
* Well populations random or fixed
* If random, minimum and maximum population sizes
* If fixed
* Number of sections on plate
* Number of T cells per well
* per section, if more than one section
* Dropout rate
Files are in CSV format. There are no header labels. Every row represents a well.
Every column represents an individual cell, containing four sequences, depicted as an array string:
Every value represents an individual cell, containing four sequences, depicted as an array string:
`[CDR3A, CDR3B, CDR1A, CDR1B]`. So a representative cell might look like this:
`[525902, 791533, -1, 866282]`
@@ -139,7 +173,6 @@ Dropout sequences are replaced with the value `-1`. Comments are preceded by `#`
Structure:
---
```
# Cell source file name:
# Each row represents one well on the plate
@@ -155,8 +188,8 @@ Structure:
---
#### Graph and Data Files
Graph and Data files are serialized binaries of a Java object containing the weigthed bipartite graph representation of a
#### Graph/Data Files
Graph/Data files are serialized binaries of a Java object containing the weigthed bipartite graph representation of a
Sample Plate, along with the necessary metadata for matching and results output. Making them requires a Cell Sample file
(to construct a list of correct sequence pairs for checking the accuracy of BiGpairSEQ simulations) and a
Sample Plate file (to construct the associated occupancy graph).
@@ -164,22 +197,23 @@ Sample Plate file (to construct the associated occupancy graph).
These files can be several gigabytes in size. Writing them to a file lets us generate a graph and its metadata once,
then use it for multiple different BiGpairSEQ simulations.
Options for creating a Graph and Data file:
Options for creating a Graph/Data file:
* The Cell Sample file to use
* The Sample Plate file to use. (This must have been generated from the selected Cell Sample file.)
These files do not have a human-readable structure, and are not portable to other programs. (Export of graphs in a
portable data format may be implemented in the future. The tricky part is encoding the necessary metadata.)
These files do not have a human-readable structure, and are not portable to other programs.
(For portability to other software, turn on GraphML output in the Options menu. This will produce a .graphml file
for the weighted graph, with vertex attributes sequence, type, and occupancy data.)
---
#### Matching Results Files
Matching results files consist of the results of a BiGpairSEQ matching simulation. Making them requires a Graph and
Data file. To save file I/O time, the data from the most recent Graph and Data file read or generated is cached
by the simulator. Subsequent BiGpairSEQ simulations run with the same input filename will use the cached version
rather than reading in again from disk.
Matching results files consist of the results of a BiGpairSEQ matching simulation. Making them requires a serialized
binary Graph/Data file (.ser). (Because .graphML files are larger than .ser files, BiGpairSEQ_Sim supports .graphML
output only. Graph/data input must use a serialized binary.)
Files are in CSV format. Rows are sequence pairings with extra relevant data. Columns are pairing-specific details.
Matching results files are in CSV format. Rows are sequence pairings with extra relevant data. Columns are pairing-specific details.
Metadata about the matching simulation is included as comments. Comments are preceded by `#`.
Options when running a BiGpairSEQ simulation of CDR3 alpha/beta matching:
@@ -194,7 +228,6 @@ Options when running a BiGpairSEQ simulation of CDR3 alpha/beta matching:
Example output:
---
```
# Source Sample Plate file: 4MilCellsPlate.csv
# Source Graph and Data file: 4MilCellsPlateGraph.ser
@@ -245,27 +278,31 @@ slightly less time than the simulation itself. Real elapsed time from start to f
## TODO
* ~~Try invoking GC at end of workloads to reduce paging to disk~~ DONE
* Hold graph data in memory until another graph is read-in? ~~ABANDONED~~ ~~UNABANDONED~~ DONE
* ~~Hold graph data in memory until another graph is read-in? ABANDONED UNABANDONED~~ DONE
* ~~*No, this won't work, because BiGpairSEQ simulations alter the underlying graph based on filtering constraints. Changes would cascade with multiple experiments.*~~
* Might have figured out a way to do it, by taking edges out and then putting them back into the graph. This may actually be possible. If so, awesome.
* Might have figured out a way to do it, by taking edges out and then putting them back into the graph. This may actually be possible.
* It is possible, though the modifications to the graph incur their own performance penalties. Need testing to see which option is best.
* ~~Test whether pairing heap (currently used) or Fibonacci heap is more efficient for priority queue in current matching algorithm~~ DONE
* ~~in theory Fibonacci heap should be more efficient, but complexity overhead may eliminate theoretical advantage~~
* ~~Add controllable heap-type parameter?~~
* Parameter implemented. Fibonacci heap the current default.
* ~~Implement sample plates with random numbers of T cells per well.~~ DONE
* Possible BiGpairSEQ advantage over pairSEQ: BiGpairSEQ is resilient to variations in well population sizes on a sample plate; pairSEQ is not.
* preliminary data suggests that BiGpairSEQ behaves roughly as though the whole plate had whatever the *average* well concentration is, but that's still speculative.
* See if there's a reasonable way to reformat Sample Plate files so that wells are columns instead of rows.
* ~~Problem is variable number of cells in a well~~
* ~~Apache Commons CSV library writes entries a row at a time~~
* _Got this working, but at the cost of a profoundly strange bug in graph occupancy filtering. Have reverted the repo until I can figure out what caused that. Given how easily Thingiverse transposes CSV matrices in R, might not even be worth fixing._
* Re-implement command line arguments, to enable scripting and statistical simulation studies
* Implement sample plates with random numbers of T cells per well.
* Possible BiGpairSEQ advantage over pairSEQ: BiGpairSEQ is resilient to variations in well population sizes on a sample plate; pairSEQ is not.
* preliminary data suggests that BiGpairSEQ behaves roughly as though the whole plate had whatever the *average* well concentration is, but that's still speculative.
* Enable GraphML output in addition to serialized object binaries, for data portability
* Custom vertex type with attribute for sequence occupancy?
* _Got this working, but at the cost of a profoundly strange bug in graph occupancy filtering. Have reverted the repo until I can figure out what caused that. Given how easily Thingiverse transposes CSV matrices in R, might not even be worth fixing.
* ~~Enable GraphML output in addition to serialized object binaries, for data portability~~ DONE
* ~~Custom vertex type with attribute for sequence occupancy?~~ ABANDONED
* Have a branch where this is implemented, but there's a bug that broke matching. Don't currently have time to fix.
* ~~Re-implement command line arguments, to enable scripting and statistical simulation studies~~ DONE
* Re-implement CDR1 matching method
* Implement Duan and Su's maximum weight matching algorithm
* Add controllable algorithm-type parameter?
* Test whether pairing heap (currently used) or Fibonacci heap is more efficient for priority queue in current matching algorithm
* in theory Fibonacci heap should be more efficient, but complexity overhead may eliminate theoretical advantage
* Add controllable heap-type parameter?
* Add controllable algorithm-type parameter?
* This would be fun and valuable, but probably take more time than I have for a hobby project.
* Implement Vose's alias method for arbitrary statistical distributions of cells
## CITATIONS
* Howie, B., Sherwood, A. M., et al. ["High-throughput pairing of T cell receptor alpha and beta sequences."](https://pubmed.ncbi.nlm.nih.gov/26290413/) Sci. Transl. Med. 7, 301ra131 (2015)

View File

@@ -1,8 +1,21 @@
//main class. Only job is to choose which interface to use, and hold graph data in memory
import java.util.Random;
//main class. For choosing interface type and holding settings
public class BiGpairSEQ {
private static final Random rand = new Random();
private static CellSample cellSampleInMemory = null;
private static String cellFilename = null;
private static Plate plateInMemory = null;
private static String plateFilename = null;
private static GraphWithMapData graphInMemory = null;
private static String graphFilename = null;
private static boolean cacheCells = false;
private static boolean cachePlate = false;
private static boolean cacheGraph = false;
private static String priorityQueueHeapType = "FIBONACCI";
private static boolean outputBinary = true;
private static boolean outputGraphML = false;
public static void main(String[] args) {
if (args.length == 0) {
@@ -10,33 +23,154 @@ public class BiGpairSEQ {
}
else {
//This will be uncommented when command line arguments are re-implemented.
//CommandLineInterface.startCLI(args);
System.out.println("Command line arguments are still being re-implemented.");
CommandLineInterface.startCLI(args);
//System.out.println("Command line arguments are still being re-implemented.");
}
}
public static GraphWithMapData getGraph() {
return graphInMemory;
public static Random getRand() {
return rand;
}
public static void setGraph(GraphWithMapData g) {
public static CellSample getCellSampleInMemory() {
return cellSampleInMemory;
}
public static void setCellSampleInMemory(CellSample cellSample, String filename) {
if(cellSampleInMemory != null) {
clearCellSampleInMemory();
}
cellSampleInMemory = cellSample;
cellFilename = filename;
System.out.println("Cell sample file " + filename + " cached.");
}
public static void clearCellSampleInMemory() {
cellSampleInMemory = null;
cellFilename = null;
System.gc();
System.out.println("Cell sample file cache cleared.");
}
public static String getCellFilename() {
return cellFilename;
}
public static Plate getPlateInMemory() {
return plateInMemory;
}
public static void setPlateInMemory(Plate plate, String filename) {
if(plateInMemory != null) {
clearPlateInMemory();
}
plateInMemory = plate;
plateFilename = filename;
System.out.println("Sample plate file " + filename + " cached.");
}
public static void clearPlateInMemory() {
plateInMemory = null;
plateFilename = null;
System.gc();
System.out.println("Sample plate file cache cleared.");
}
public static String getPlateFilename() {
return plateFilename;
}
public static GraphWithMapData getGraphInMemory() {return graphInMemory;
}
public static void setGraphInMemory(GraphWithMapData g, String filename) {
if (graphInMemory != null) {
clearGraph();
clearGraphInMemory();
}
graphInMemory = g;
graphFilename = filename;
System.out.println("Graph and data file " + filename + " cached.");
}
public static void clearGraph() {
public static void clearGraphInMemory() {
graphInMemory = null;
graphFilename = null;
System.gc();
System.out.println("Graph and data file cache cleared.");
}
public static String getGraphFilename() {
return graphFilename;
}
public static void setGraphFilename(String filename) {
graphFilename = filename;
public static boolean cacheCells() {
return cacheCells;
}
public static void setCacheCells(boolean cacheCells) {
//if not caching, clear the memory
if(!cacheCells){
BiGpairSEQ.clearCellSampleInMemory();
System.out.println("Cell sample file caching: OFF.");
}
else {
System.out.println("Cell sample file caching: ON.");
}
BiGpairSEQ.cacheCells = cacheCells;
}
public static boolean cachePlate() {
return cachePlate;
}
public static void setCachePlate(boolean cachePlate) {
//if not caching, clear the memory
if(!cachePlate) {
BiGpairSEQ.clearPlateInMemory();
System.out.println("Sample plate file caching: OFF.");
}
else {
System.out.println("Sample plate file caching: ON.");
}
BiGpairSEQ.cachePlate = cachePlate;
}
public static boolean cacheGraph() {
return cacheGraph;
}
public static void setCacheGraph(boolean cacheGraph) {
//if not caching, clear the memory
if(!cacheGraph) {
BiGpairSEQ.clearGraphInMemory();
System.out.println("Graph/data file caching: OFF.");
}
else {
System.out.println("Graph/data file caching: ON.");
}
BiGpairSEQ.cacheGraph = cacheGraph;
}
public static String getPriorityQueueHeapType() {
return priorityQueueHeapType;
}
public static void setPairingHeap() {
priorityQueueHeapType = "PAIRING";
}
public static void setFibonacciHeap() {
priorityQueueHeapType = "FIBONACCI";
}
public static boolean outputBinary() {return outputBinary;}
public static void setOutputBinary(boolean b) {outputBinary = b;}
public static boolean outputGraphML() {return outputGraphML;}
public static void setOutputGraphML(boolean b) {outputGraphML = b;}
}

View File

@@ -13,6 +13,7 @@ public class CellFileReader {
private String filename;
private List<Integer[]> distinctCells = new ArrayList<>();
private Integer cdr1Freq;
public CellFileReader(String filename) {
if(!filename.matches(".*\\.csv")){
@@ -38,19 +39,37 @@ public class CellFileReader {
cell[3] = Integer.valueOf(record.get("Beta CDR1"));
distinctCells.add(cell);
}
} catch(IOException ex){
System.out.println("cell file " + filename + " not found.");
System.err.println(ex);
}
//get CDR1 frequency
ArrayList<Integer> cdr1Alphas = new ArrayList<>();
for (Integer[] cell : distinctCells) {
cdr1Alphas.add(cell[3]);
}
double count = cdr1Alphas.stream().distinct().count();
count = Math.ceil(distinctCells.size() / count);
cdr1Freq = (int) count;
}
public CellSample getCellSample() {
return new CellSample(distinctCells, cdr1Freq);
}
public String getFilename() { return filename;}
public List<Integer[]> getCells(){
//Refactor everything that uses this to have access to a Cell Sample and get the cells there instead.
public List<Integer[]> getListOfDistinctCellsDEPRECATED(){
return distinctCells;
}
public Integer getCellCount() {
public Integer getCellCountDEPRECATED() {
//Refactor everything that uses this to have access to a Cell Sample and get the count there instead.
return distinctCells.size();
}
}

View File

@@ -1,10 +1,37 @@
import java.util.ArrayList;
import java.util.Collections;
import java.util.List;
import java.util.stream.IntStream;
public class CellSample {
private List<Integer[]> cells;
private Integer cdr1Freq;
public CellSample(Integer numDistinctCells, Integer cdr1Freq){
this.cdr1Freq = cdr1Freq;
List<Integer> numbersCDR3 = new ArrayList<>();
List<Integer> numbersCDR1 = new ArrayList<>();
Integer numDistCDR3s = 2 * numDistinctCells + 1;
IntStream.range(1, numDistCDR3s + 1).forEach(i -> numbersCDR3.add(i));
IntStream.range(numDistCDR3s + 1, numDistCDR3s + 1 + (numDistCDR3s / cdr1Freq) + 1).forEach(i -> numbersCDR1.add(i));
Collections.shuffle(numbersCDR3);
Collections.shuffle(numbersCDR1);
//Each cell represented by 4 values
//two CDR3s, and two CDR1s. First two values are CDR3s (alpha, beta), second two are CDR1s (alpha, beta)
List<Integer[]> distinctCells = new ArrayList<>();
for(int i = 0; i < numbersCDR3.size() - 1; i = i + 2){
Integer tmpCDR3a = numbersCDR3.get(i);
Integer tmpCDR3b = numbersCDR3.get(i+1);
Integer tmpCDR1a = numbersCDR1.get(i % numbersCDR1.size());
Integer tmpCDR1b = numbersCDR1.get((i+1) % numbersCDR1.size());
Integer[] tmp = {tmpCDR3a, tmpCDR3b, tmpCDR1a, tmpCDR1b};
distinctCells.add(tmp);
}
this.cells = distinctCells;
}
public CellSample(List<Integer[]> cells, Integer cdr1Freq){
this.cells = cells;
this.cdr1Freq = cdr1Freq;
@@ -18,7 +45,7 @@ public class CellSample {
return cdr1Freq;
}
public Integer population(){
public Integer getCellCount(){
return cells.size();
}

View File

@@ -1,5 +1,9 @@
import org.apache.commons.cli.*;
import java.io.IOException;
import java.util.Arrays;
import java.util.stream.Stream;
/*
* Class for parsing options passed to program from command line
*
@@ -29,6 +33,8 @@ import org.apache.commons.cli.*;
* cellfile : name of the cell sample file to use as input
* platefile : name of the sample plate file to use as input
* output : name of the output file
* graphml : output a graphml file
* binary : output a serialized binary object file
*
* Match flags:
* graphFile : name of graph and data file to use as input
@@ -43,286 +49,379 @@ import org.apache.commons.cli.*;
public class CommandLineInterface {
public static void startCLI(String[] args) {
//These command line options are a big mess
//Really, I don't think command line tools are expected to work in this many different modes
//making cells, making plates, and matching are the sort of thing that UNIX philosophy would say
//should be three separate programs.
//There might be a way to do it with option parameters?
//main options set
Options mainOptions = new Options();
Option makeCells = Option.builder("cells")
.longOpt("make-cells")
.desc("Makes a file of distinct cells")
.build();
Option makePlate = Option.builder("plates")
.longOpt("make-plates")
.desc("Makes a sample plate file")
.build();
Option makeGraph = Option.builder("graph")
.longOpt("make-graph")
.desc("Makes a graph and data file")
.build();
Option matchCDR3 = Option.builder("match")
.longOpt("match-cdr3")
.desc("Match CDR3s. Requires a cell sample file and any number of plate files.")
.build();
OptionGroup mainGroup = new OptionGroup();
mainGroup.addOption(makeCells);
mainGroup.addOption(makePlate);
mainGroup.addOption(makeGraph);
mainGroup.addOption(matchCDR3);
mainGroup.setRequired(true);
mainOptions.addOptionGroup(mainGroup);
//Reuse clones of this for other options groups, rather than making it lots of times
Option outputFile = Option.builder("o")
.longOpt("output-file")
.hasArg()
.argName("filename")
.desc("Name of output file")
.build();
mainOptions.addOption(outputFile);
//Options cellOptions = new Options();
Option numCells = Option.builder("nc")
.longOpt("num-cells")
.desc("The number of distinct cells to generate")
.hasArg()
.argName("number")
.build();
mainOptions.addOption(numCells);
Option cdr1Freq = Option.builder("d")
.longOpt("peptide-diversity-factor")
.hasArg()
.argName("number")
.desc("Number of distinct CDR3s for every CDR1")
.build();
mainOptions.addOption(cdr1Freq);
//Option cellOutput = (Option) outputFile.clone();
//cellOutput.setRequired(true);
//mainOptions.addOption(cellOutput);
//Options plateOptions = new Options();
Option inputCells = Option.builder("c")
.longOpt("cell-file")
.hasArg()
.argName("file")
.desc("The cell sample file used for filling wells")
.build();
mainOptions.addOption(inputCells);
Option numWells = Option.builder("w")
.longOpt("num-wells")
.hasArg()
.argName("number")
.desc("The number of wells on each plate")
.build();
mainOptions.addOption(numWells);
Option numPlates = Option.builder("np")
.longOpt("num-plates")
.hasArg()
.argName("number")
.desc("The number of plate files to output")
.build();
mainOptions.addOption(numPlates);
//Option plateOutput = (Option) outputFile.clone();
//plateOutput.setRequired(true);
//plateOutput.setDescription("Prefix for plate output filenames");
//mainOptions.addOption(plateOutput);
Option plateErr = Option.builder("err")
.longOpt("drop-out-rate")
.hasArg()
.argName("number")
.desc("Well drop-out rate. (Probability between 0 and 1)")
.build();
mainOptions.addOption(plateErr);
Option plateConcentrations = Option.builder("t")
.longOpt("t-cells-per-well")
.hasArgs()
.argName("number 1, number 2, ...")
.desc("Number of T cells per well for each plate section")
.build();
mainOptions.addOption(plateConcentrations);
//different distributions, mutually exclusive
OptionGroup plateDistributions = new OptionGroup();
Option plateExp = Option.builder("exponential")
.desc("Sample from distinct cells with exponential frequency distribution")
.build();
plateDistributions.addOption(plateExp);
Option plateGaussian = Option.builder("gaussian")
.desc("Sample from distinct cells with gaussain frequency distribution")
.build();
plateDistributions.addOption(plateGaussian);
Option platePoisson = Option.builder("poisson")
.desc("Sample from distinct cells with poisson frequency distribution")
.build();
plateDistributions.addOption(platePoisson);
mainOptions.addOptionGroup(plateDistributions);
Option plateStdDev = Option.builder("stddev")
.desc("Standard deviation for gaussian distribution")
.hasArg()
.argName("number")
.build();
mainOptions.addOption(plateStdDev);
Option plateLambda = Option.builder("lambda")
.desc("Lambda for exponential distribution")
.hasArg()
.argName("number")
.build();
mainOptions.addOption(plateLambda);
//
// String cellFile, String filename, Double stdDev,
// Integer numWells, Integer numSections,
// Integer[] concentrations, Double dropOutRate
//
//Options matchOptions = new Options();
inputCells.setDescription("The cell sample file to be used for matching.");
mainOptions.addOption(inputCells);
Option lowThresh = Option.builder("low")
.longOpt("low-threshold")
.hasArg()
.argName("number")
.desc("Sets the minimum occupancy overlap to attempt matching")
.build();
mainOptions.addOption(lowThresh);
Option highThresh = Option.builder("high")
.longOpt("high-threshold")
.hasArg()
.argName("number")
.desc("Sets the maximum occupancy overlap to attempt matching")
.build();
mainOptions.addOption(highThresh);
Option occDiff = Option.builder("occdiff")
.longOpt("occupancy-difference")
.hasArg()
.argName("Number")
.desc("Maximum difference in alpha/beta occupancy to attempt matching")
.build();
mainOptions.addOption(occDiff);
Option overlapPer = Option.builder("ovper")
.longOpt("overlap-percent")
.hasArg()
.argName("Percent")
.desc("Minimum overlap percent to attempt matching (0 -100)")
.build();
mainOptions.addOption(overlapPer);
Option inputPlates = Option.builder("p")
.longOpt("plate-files")
.hasArgs()
.desc("Plate files to match")
.build();
mainOptions.addOption(inputPlates);
//Options sets for the different modes
Options mainOptions = buildMainOptions();
Options cellOptions = buildCellOptions();
Options plateOptions = buildPlateOptions();
Options graphOptions = buildGraphOptions();
Options matchOptions = buildMatchCDR3options();
CommandLineParser parser = new DefaultParser();
try {
CommandLine line = parser.parse(mainOptions, args);
if(line.hasOption("match")){
//line = parser.parse(mainOptions, args);
//String cellFile = line.getOptionValue("c");
String graphFile = line.getOptionValue("g");
Integer lowThreshold = Integer.valueOf(line.getOptionValue(lowThresh));
Integer highThreshold = Integer.valueOf(line.getOptionValue(highThresh));
Integer occupancyDifference = Integer.valueOf(line.getOptionValue(occDiff));
Integer overlapPercent = Integer.valueOf(line.getOptionValue(overlapPer));
for(String plate: line.getOptionValues("p")) {
matchCDR3s(graphFile, lowThreshold, highThreshold, occupancyDifference, overlapPercent);
}
try{
CommandLine line = parser.parse(mainOptions, Arrays.copyOfRange(args, 0, 1));
if (line.hasOption("help")) {
HelpFormatter formatter = new HelpFormatter();
formatter.printHelp("BiGpairSEQ_Sim", mainOptions);
System.out.println();
formatter.printHelp("BiGpairSEQ_SIM -cells", cellOptions);
System.out.println();
formatter.printHelp("BiGpairSEQ_Sim -plate", plateOptions);
System.out.println();
formatter.printHelp("BiGpairSEQ_Sim -graph", graphOptions);
System.out.println();
formatter.printHelp("BiGpairSEQ_Sim -match", matchOptions);
}
else if(line.hasOption("cells")){
//line = parser.parse(mainOptions, args);
else if (line.hasOption("cells")) {
line = parser.parse(cellOptions, Arrays.copyOfRange(args, 1, args.length));
Integer number = Integer.valueOf(line.getOptionValue("n"));
Integer diversity = Integer.valueOf(line.getOptionValue("d"));
String filename = line.getOptionValue("o");
Integer numDistCells = Integer.valueOf(line.getOptionValue("nc"));
Integer freq = Integer.valueOf(line.getOptionValue("d"));
makeCells(filename, numDistCells, freq);
makeCells(filename, number, diversity);
}
else if(line.hasOption("plates")){
//line = parser.parse(mainOptions, args);
String cellFile = line.getOptionValue("c");
String filenamePrefix = line.getOptionValue("o");
Integer numWellsOnPlate = Integer.valueOf(line.getOptionValue("w"));
Integer numPlatesToMake = Integer.valueOf(line.getOptionValue("np"));
String[] concentrationsToUseString = line.getOptionValues("t");
Integer numSections = concentrationsToUseString.length;
Integer[] concentrationsToUse = new Integer[numSections];
for(int i = 0; i <numSections; i++){
concentrationsToUse[i] = Integer.valueOf(concentrationsToUseString[i]);
else if (line.hasOption("plate")) {
line = parser.parse(plateOptions, Arrays.copyOfRange(args, 1, args.length));
//get the cells
String cellFilename = line.getOptionValue("c");
CellSample cells = getCells(cellFilename);
//get the rest of the parameters
Integer[] populations;
String outputFilename = line.getOptionValue("o");
Integer numWells = Integer.parseInt(line.getOptionValue("w"));
Double dropoutRate = Double.parseDouble(line.getOptionValue("err"));
if (line.hasOption("random")) {
//Array holding values of minimum and maximum populations
Integer[] min_max = Stream.of(line.getOptionValues("random"))
.mapToInt(Integer::parseInt)
.boxed()
.toArray(Integer[]::new);
populations = BiGpairSEQ.getRand().ints(min_max[0], min_max[1] + 1)
.limit(numWells)
.boxed()
.toArray(Integer[]::new);
}
Double dropOutRate = Double.valueOf(line.getOptionValue("err"));
if(line.hasOption("exponential")){
Double lambda = Double.valueOf(line.getOptionValue("lambda"));
for(int i = 1; i <= numPlatesToMake; i++){
makePlateExp(cellFile, filenamePrefix + i, lambda, numWellsOnPlate,
concentrationsToUse,dropOutRate);
}
else if (line.hasOption("pop")) {
populations = Stream.of(line.getOptionValues("pop"))
.mapToInt(Integer::parseInt)
.boxed()
.toArray(Integer[]::new);
}
else if(line.hasOption("gaussian")){
Double stdDev = Double.valueOf(line.getOptionValue("std-dev"));
for(int i = 1; i <= numPlatesToMake; i++){
makePlate(cellFile, filenamePrefix + i, stdDev, numWellsOnPlate,
concentrationsToUse,dropOutRate);
}
else{
populations = new Integer[1];
populations[0] = 1;
}
//make the plate
Plate plate;
if (line.hasOption("poisson")) {
Double stdDev = Math.sqrt(numWells);
plate = new Plate(cells, cellFilename, numWells, populations, dropoutRate, stdDev, false);
}
else if (line.hasOption("gaussian")) {
Double stdDev = Double.parseDouble(line.getOptionValue("stddev"));
plate = new Plate(cells, cellFilename, numWells, populations, dropoutRate, stdDev, false);
}
else {
assert line.hasOption("exponential");
Double lambda = Double.parseDouble(line.getOptionValue("lambda"));
plate = new Plate(cells, cellFilename, numWells, populations, dropoutRate, lambda, true);
}
PlateFileWriter writer = new PlateFileWriter(outputFilename, plate);
writer.writePlateFile();
}
else if (line.hasOption("graph")) { //Making a graph
line = parser.parse(graphOptions, Arrays.copyOfRange(args, 1, args.length));
String cellFilename = line.getOptionValue("c");
String plateFilename = line.getOptionValue("p");
String outputFilename = line.getOptionValue("o");
//get cells
CellSample cells = getCells(cellFilename);
//get plate
Plate plate = getPlate(plateFilename);
GraphWithMapData graph = Simulator.makeGraph(cells, plate, false);
if (!line.hasOption("no-binary")) { //output binary file unless told not to
GraphDataObjectWriter writer = new GraphDataObjectWriter(outputFilename, graph, false);
writer.writeDataToFile();
}
else if(line.hasOption("poisson")){
for(int i = 1; i <= numPlatesToMake; i++){
makePlatePoisson(cellFile, filenamePrefix + i, numWellsOnPlate,
concentrationsToUse,dropOutRate);
}
if (line.hasOption("graphml")) { //if told to, output graphml file
GraphMLFileWriter gmlwriter = new GraphMLFileWriter(outputFilename, graph);
gmlwriter.writeGraphToFile();
}
}
else if (line.hasOption("match")) { //can add a flag for which match type in future, spit this in two
line = parser.parse(matchOptions, Arrays.copyOfRange(args, 1, args.length));
String graphFilename = line.getOptionValue("g");
String outputFilename = line.getOptionValue("o");
Integer minThreshold = Integer.parseInt(line.getOptionValue("min"));
Integer maxThreshold = Integer.parseInt(line.getOptionValue("max"));
Integer minOverlapPct;
if (line.hasOption("minpct")) { //see if this filter is being used
minOverlapPct = Integer.parseInt(line.getOptionValue("minpct"));
}
else {
minOverlapPct = 0;
}
Integer maxOccupancyDiff;
if (line.hasOption("maxdiff")) { //see if this filter is being used
maxOccupancyDiff = Integer.parseInt(line.getOptionValue("maxdiff"));
}
else {
maxOccupancyDiff = Integer.MAX_VALUE;
}
GraphWithMapData graph = getGraph(graphFilename);
MatchingResult result = Simulator.matchCDR3s(graph, graphFilename, minThreshold, maxThreshold,
maxOccupancyDiff, minOverlapPct, false);
MatchingFileWriter writer = new MatchingFileWriter(outputFilename, result);
writer.writeResultsToFile();
//can put a bunch of ifs for outputting various things from the MatchingResult to System.out here
//after I put those flags in the matchOptions
}
}
catch (ParseException exp) {
System.err.println("Parsing failed. Reason: " + exp.getMessage());
}
}
private static Option outputFileOption() {
Option outputFile = Option.builder("o")
.longOpt("output-file")
.hasArg()
.argName("filename")
.desc("Name of output file")
.required()
.build();
return outputFile;
}
private static Options buildMainOptions() {
Options mainOptions = new Options();
Option help = Option.builder("help")
.desc("Displays this help menu")
.build();
Option makeCells = Option.builder("cells")
.longOpt("make-cells")
.desc("Makes a cell sample file of distinct T cells")
.build();
Option makePlate = Option.builder("plate")
.longOpt("make-plate")
.desc("Makes a sample plate file. Requires a cell sample file.")
.build();
Option makeGraph = Option.builder("graph")
.longOpt("make-graph")
.desc("Makes a graph/data file. Requires a cell sample file and a sample plate file")
.build();
Option matchCDR3 = Option.builder("match")
.longOpt("match-cdr3")
.desc("Matches CDR3s. Requires a graph/data file.")
.build();
OptionGroup mainGroup = new OptionGroup();
mainGroup.addOption(help);
mainGroup.addOption(makeCells);
mainGroup.addOption(makePlate);
mainGroup.addOption(makeGraph);
mainGroup.addOption(matchCDR3);
mainGroup.setRequired(true);
mainOptions.addOptionGroup(mainGroup);
return mainOptions;
}
private static Options buildCellOptions() {
Options cellOptions = new Options();
Option numCells = Option.builder("n")
.longOpt("num-cells")
.desc("The number of distinct cells to generate")
.hasArg()
.argName("number")
.required().build();
Option cdr3Diversity = Option.builder("d")
.longOpt("diversity-factor")
.desc("The factor by which unique CDR3s outnumber unique CDR1s")
.hasArg()
.argName("factor")
.required().build();
cellOptions.addOption(numCells);
cellOptions.addOption(cdr3Diversity);
cellOptions.addOption(outputFileOption());
return cellOptions;
}
private static Options buildPlateOptions() {
Options plateOptions = new Options();
Option cellFile = Option.builder("c") // add this to plate options
.longOpt("cell-file")
.desc("The cell sample file to use")
.hasArg()
.argName("filename")
.required().build();
Option numWells = Option.builder("w")// add this to plate options
.longOpt("wells")
.desc("The number of wells on the sample plate")
.hasArg()
.argName("number")
.required().build();
//options group for choosing with distribution to use
OptionGroup distributions = new OptionGroup();// add this to plate options
distributions.setRequired(true);
Option poisson = Option.builder("poisson")
.desc("Use a Poisson distribution for cell sample")
.build();
Option gaussian = Option.builder("gaussian")
.desc("Use a Gaussian distribution for cell sample")
.build();
Option exponential = Option.builder("exponential")
.desc("Use an exponential distribution for cell sample")
.build();
distributions.addOption(poisson);
distributions.addOption(gaussian);
distributions.addOption(exponential);
//options group for statistical distribution parameters
OptionGroup statParams = new OptionGroup();// add this to plate options
Option stdDev = Option.builder("stddev")
.desc("If using -gaussian flag, standard deviation for distrbution")
.hasArg()
.argName("value")
.build();
Option lambda = Option.builder("lambda")
.desc("If using -exponential flag, lambda value for distribution")
.hasArg()
.argName("value")
.build();
statParams.addOption(stdDev);
statParams.addOption(lambda);
//Option group for random plate or set populations
OptionGroup wellPopOptions = new OptionGroup(); // add this to plate options
wellPopOptions.setRequired(true);
Option randomWellPopulations = Option.builder("random")
.desc("Randomize well populations on sample plate. Takes two arguments: the minimum possible population and the maximum possible population.")
.hasArgs()
.numberOfArgs(2)
.argName("minimum maximum")
.build();
Option specificWellPopulations = Option.builder("pop")
.desc("The well populations for each section of the sample plate. There will be as many sections as there are populations given.")
.hasArgs()
.argName("number [number]...")
.build();
Option dropoutRate = Option.builder("err") //add this to plate options
.hasArg()
.desc("The sequence dropout rate due to amplification error. (0.0 - 1.0)")
.argName("rate")
.required()
.build();
wellPopOptions.addOption(randomWellPopulations);
wellPopOptions.addOption(specificWellPopulations);
plateOptions.addOption(cellFile);
plateOptions.addOption(numWells);
plateOptions.addOptionGroup(distributions);
plateOptions.addOptionGroup(statParams);
plateOptions.addOptionGroup(wellPopOptions);
plateOptions.addOption(dropoutRate);
plateOptions.addOption(outputFileOption());
return plateOptions;
}
private static Options buildGraphOptions() {
Options graphOptions = new Options();
Option cellFilename = Option.builder("c")
.longOpt("cell-file")
.desc("Cell sample file to use for checking accuracy")
.hasArg()
.argName("filename")
.required().build();
Option plateFilename = Option.builder("p")
.longOpt("plate-filename")
.desc("Sample plate file (made from given cell sample file) to construct graph from")
.hasArg()
.argName("filename")
.required().build();
Option outputGraphML = Option.builder("graphml")
.desc("Output GraphML file")
.build();
Option outputSerializedBinary = Option.builder("nb")
.longOpt("no-binary")
.desc("Don't output serialized binary file")
.build();
graphOptions.addOption(cellFilename);
graphOptions.addOption(plateFilename);
graphOptions.addOption(outputFileOption());
graphOptions.addOption(outputGraphML);
graphOptions.addOption(outputSerializedBinary);
return graphOptions;
}
private static Options buildMatchCDR3options() {
Options matchCDR3options = new Options();
Option graphFilename = Option.builder("g")
.longOpt("graph-file")
.desc("The graph/data file to use")
.hasArg()
.argName("filename")
.required().build();
Option minOccupancyOverlap = Option.builder("min")
.desc("The minimum number of shared wells to attempt to match a sequence pair")
.hasArg()
.argName("number")
.required().build();
Option maxOccupancyOverlap = Option.builder("max")
.desc("The maximum number of shared wells to attempt to match a sequence pair")
.hasArg()
.argName("number")
.required().build();
Option minOverlapPercent = Option.builder("minpct")
.desc("(Optional) The minimum percentage of a sequence's total occupancy shared by another sequence to attempt matching. (0 - 100) ")
.hasArg()
.argName("percent")
.build();
Option maxOccupancyDifference = Option.builder("maxdiff")
.desc("(Optional) The maximum difference in total occupancy between two sequences to attempt matching.")
.hasArg()
.argName("number")
.build();
matchCDR3options.addOption(graphFilename);
matchCDR3options.addOption(minOccupancyOverlap);
matchCDR3options.addOption(maxOccupancyOverlap);
matchCDR3options.addOption(minOverlapPercent);
matchCDR3options.addOption(maxOccupancyDifference);
matchCDR3options.addOption(outputFileOption());
//options for output to System.out
//Option printPairingErrorRate = Option.builder()
return matchCDR3options;
}
private static CellSample getCells(String cellFilename) {
assert cellFilename != null;
CellFileReader reader = new CellFileReader(cellFilename);
return reader.getCellSample();
}
private static Plate getPlate(String plateFilename) {
assert plateFilename != null;
PlateFileReader reader = new PlateFileReader(plateFilename);
return reader.getSamplePlate();
}
private static GraphWithMapData getGraph(String graphFilename) {
assert graphFilename != null;
try{
GraphDataObjectReader reader = new GraphDataObjectReader(graphFilename, false);
return reader.getData();
}
catch (IOException ex) {
ex.printStackTrace();
return null;
}
}
//for calling from command line
public static void makeCells(String filename, Integer numCells, Integer cdr1Freq){
CellSample sample = Simulator.generateCellSample(numCells, cdr1Freq);
public static void makeCells(String filename, Integer numCells, Integer cdr1Freq) {
CellSample sample = new CellSample(numCells, cdr1Freq);
CellFileWriter writer = new CellFileWriter(filename, sample);
writer.writeCellsToFile();
}
public static void makePlateExp(String cellFile, String filename, Double lambda,
Integer numWells, Integer[] concentrations, Double dropOutRate){
CellFileReader cellReader = new CellFileReader(cellFile);
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWellsExponential(cellReader.getFilename(), cellReader.getCells(), lambda);
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
writer.writePlateFile();
}
private static void makePlatePoisson(String cellFile, String filename, Integer numWells,
Integer[] concentrations, Double dropOutRate){
CellFileReader cellReader = new CellFileReader(cellFile);
Double stdDev = Math.sqrt(cellReader.getCellCount());
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWells(cellReader.getFilename(), cellReader.getCells(), stdDev);
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
writer.writePlateFile();
}
private static void makePlate(String cellFile, String filename, Double stdDev,
Integer numWells, Integer[] concentrations, Double dropOutRate){
CellFileReader cellReader = new CellFileReader(cellFile);
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWells(cellReader.getFilename(), cellReader.getCells(), stdDev);
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
writer.writePlateFile();
}
private static void matchCDR3s(String graphFile, Integer lowThreshold, Integer highThreshold,
Integer occupancyDifference, Integer overlapPercent) {
}
}

View File

@@ -4,10 +4,6 @@ import java.math.MathContext;
public abstract class Equations {
public static int getRandomNumber(int min, int max) {
return (int) ((Math.random() * (max - min)) + min);
}
//pValue calculation as described in original pairSEQ paper.
//Included for comparison with original results.
//Not used by BiGpairSEQ for matching.

View File

@@ -1,10 +1,12 @@
import java.io.*;
public class GraphDataObjectReader {
private GraphWithMapData data;
private String filename;
private boolean verbose = true;
public GraphDataObjectReader(String filename) throws IOException {
public GraphDataObjectReader(String filename, boolean verbose) throws IOException {
if(!filename.matches(".*\\.ser")){
filename = filename + ".ser";
}

View File

@@ -1,3 +1,5 @@
import org.jgrapht.Graph;
import java.io.BufferedOutputStream;
import java.io.FileOutputStream;
import java.io.IOException;
@@ -7,6 +9,7 @@ public class GraphDataObjectWriter {
private GraphWithMapData data;
private String filename;
private boolean verbose = true;
public GraphDataObjectWriter(String filename, GraphWithMapData data) {
if(!filename.matches(".*\\.ser")){
@@ -16,13 +19,24 @@ public class GraphDataObjectWriter {
this.data = data;
}
public GraphDataObjectWriter(String filename, GraphWithMapData data, boolean verbose) {
this.verbose = verbose;
if(!filename.matches(".*\\.ser")){
filename = filename + ".ser";
}
this.filename = filename;
this.data = data;
}
public void writeDataToFile() {
try (BufferedOutputStream bufferedOut = new BufferedOutputStream(new FileOutputStream(filename));
ObjectOutputStream out = new ObjectOutputStream(bufferedOut);
){
System.out.println("Writing graph and occupancy data to file. This may take some time.");
System.out.println("File I/O time is not included in results.");
if(verbose) {
System.out.println("Writing graph and occupancy data to file. This may take some time.");
System.out.println("File I/O time is not included in results.");
}
out.writeObject(data);
} catch (IOException ex) {
ex.printStackTrace();

View File

@@ -1,35 +0,0 @@
import org.jgrapht.graph.SimpleWeightedGraph;
import org.jgrapht.nio.graphml.GraphMLImporter;
import java.io.BufferedReader;
import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
public class GraphMLFileReader {
private String filename;
private SimpleWeightedGraph graph;
public GraphMLFileReader(String filename, SimpleWeightedGraph graph) {
if(!filename.matches(".*\\.graphml")){
filename = filename + ".graphml";
}
this.filename = filename;
this.graph = graph;
try(//don't need to close reader bc of try-with-resources auto-closing
BufferedReader reader = Files.newBufferedReader(Path.of(filename));
){
GraphMLImporter<SimpleWeightedGraph, BufferedReader> importer = new GraphMLImporter<>();
importer.importGraph(graph, reader);
}
catch (IOException ex) {
System.out.println("Graph file " + filename + " not found.");
System.err.println(ex);
}
}
public SimpleWeightedGraph getGraph() { return graph; }
}

View File

@@ -1,4 +1,8 @@
import org.jgrapht.graph.DefaultWeightedEdge;
import org.jgrapht.graph.SimpleWeightedGraph;
import org.jgrapht.nio.Attribute;
import org.jgrapht.nio.AttributeType;
import org.jgrapht.nio.DefaultAttribute;
import org.jgrapht.nio.dot.DOTExporter;
import org.jgrapht.nio.graphml.GraphMLExporter;
@@ -7,25 +11,69 @@ import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;
import java.nio.file.StandardOpenOption;
import java.util.HashMap;
import java.util.LinkedHashMap;
import java.util.Map;
public class GraphMLFileWriter {
String filename;
SimpleWeightedGraph graph;
GraphWithMapData data;
public GraphMLFileWriter(String filename, SimpleWeightedGraph graph) {
public GraphMLFileWriter(String filename, GraphWithMapData data) {
if(!filename.matches(".*\\.graphml")){
filename = filename + ".graphml";
}
this.filename = filename;
this.graph = graph;
this.data = data;
}
// public void writeGraphToFile() {
// try(BufferedWriter writer = Files.newBufferedWriter(Path.of(filename), StandardOpenOption.CREATE_NEW);
// ){
// GraphMLExporter<SimpleWeightedGraph, BufferedWriter> exporter = new GraphMLExporter<>();
// exporter.exportGraph(graph, writer);
// } catch(IOException ex){
// System.out.println("Could not make new file named "+filename);
// System.err.println(ex);
// }
// }
public void writeGraphToFile() {
SimpleWeightedGraph graph = data.getGraph();
Map<Integer, Integer> vertexToAlphaMap = data.getPlateVtoAMap();
Map<Integer, Integer> vertexToBetaMap = data.getPlateVtoBMap();
Map<Integer, Integer> alphaOccs = data.getAlphaWellCounts();
Map<Integer, Integer> betaOccs = data.getBetaWellCounts();
try(BufferedWriter writer = Files.newBufferedWriter(Path.of(filename), StandardOpenOption.CREATE_NEW);
){
GraphMLExporter<SimpleWeightedGraph, BufferedWriter> exporter = new GraphMLExporter<>();
//create exporter. Let the vertex labels be the unique ids for the vertices
GraphMLExporter<Integer, SimpleWeightedGraph<Vertex, DefaultWeightedEdge>> exporter = new GraphMLExporter<>(v -> v.toString());
//set to export weights
exporter.setExportEdgeWeights(true);
//set type, sequence, and occupancy attributes for each vertex
exporter.setVertexAttributeProvider( v -> {
Map<String, Attribute> attributes = new HashMap<>();
if(vertexToAlphaMap.containsKey(v)) {
attributes.put("type", DefaultAttribute.createAttribute("CDR3 Alpha"));
attributes.put("sequence", DefaultAttribute.createAttribute(vertexToAlphaMap.get(v)));
attributes.put("occupancy", DefaultAttribute.createAttribute(
alphaOccs.get(vertexToAlphaMap.get(v))));
}
else if(vertexToBetaMap.containsKey(v)) {
attributes.put("type", DefaultAttribute.createAttribute("CDR3 Beta"));
attributes.put("sequence", DefaultAttribute.createAttribute(vertexToBetaMap.get(v)));
attributes.put("occupancy", DefaultAttribute.createAttribute(
betaOccs.get(vertexToBetaMap.get(v))));
}
return attributes;
});
//register the attributes
exporter.registerAttribute("type", GraphMLExporter.AttributeCategory.NODE, AttributeType.STRING);
exporter.registerAttribute("sequence", GraphMLExporter.AttributeCategory.NODE, AttributeType.STRING);
exporter.registerAttribute("occupancy", GraphMLExporter.AttributeCategory.NODE, AttributeType.STRING);
//export the graph
exporter.exportGraph(graph, writer);
} catch(IOException ex){
System.out.println("Could not make new file named "+filename);
@@ -33,3 +81,4 @@ public class GraphMLFileWriter {
}
}
}

View File

@@ -4,61 +4,75 @@ import org.jgrapht.graph.SimpleWeightedGraph;
import java.util.ArrayList;
import java.util.List;
import java.util.Map;
import java.util.Set;
public abstract class GraphModificationFunctions {
public interface GraphModificationFunctions {
//remove over- and under-weight edges
public static List<Integer[]> filterByOverlapThresholds(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
int low, int high) {
static List<Integer[]> filterByOverlapThresholds(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
int low, int high, boolean saveEdges) {
List<Integer[]> removedEdges = new ArrayList<>();
for(DefaultWeightedEdge e: graph.edgeSet()){
if ((graph.getEdgeWeight(e) > high) || (graph.getEdgeWeight(e) < low)){
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer weight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, weight};
removedEdges.add(edge);
for (DefaultWeightedEdge e : graph.edgeSet()) {
if ((graph.getEdgeWeight(e) > high) || (graph.getEdgeWeight(e) < low)) {
if(saveEdges) {
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer weight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, weight};
removedEdges.add(edge);
}
else {
graph.setEdgeWeight(e, 0.0);
}
}
}
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
if(saveEdges) {
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
}
}
return removedEdges;
}
//Remove edges for pairs with large occupancy discrepancy
public static List<Integer[]> filterByRelativeOccupancy(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
static List<Integer[]> filterByRelativeOccupancy(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
Map<Integer, Integer> alphaWellCounts,
Map<Integer, Integer> betaWellCounts,
Map<Integer, Integer> plateVtoAMap,
Map<Integer, Integer> plateVtoBMap,
Integer maxOccupancyDifference) {
Integer maxOccupancyDifference, boolean saveEdges) {
List<Integer[]> removedEdges = new ArrayList<>();
for (DefaultWeightedEdge e : graph.edgeSet()) {
Integer alphaOcc = alphaWellCounts.get(plateVtoAMap.get(graph.getEdgeSource(e)));
Integer betaOcc = betaWellCounts.get(plateVtoBMap.get(graph.getEdgeTarget(e)));
if (Math.abs(alphaOcc - betaOcc) >= maxOccupancyDifference) {
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer weight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, weight};
removedEdges.add(edge);
if (saveEdges) {
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer weight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, weight};
removedEdges.add(edge);
}
else {
graph.setEdgeWeight(e, 0.0);
}
}
}
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
if(saveEdges) {
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
}
}
return removedEdges;
}
//Remove edges for pairs where overlap size is significantly lower than the well occupancy
public static List<Integer[]> filterByOverlapPercent(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
static List<Integer[]> filterByOverlapPercent(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
Map<Integer, Integer> alphaWellCounts,
Map<Integer, Integer> betaWellCounts,
Map<Integer, Integer> plateVtoAMap,
Map<Integer, Integer> plateVtoBMap,
Integer minOverlapPercent) {
Integer minOverlapPercent,
boolean saveEdges) {
List<Integer[]> removedEdges = new ArrayList<>();
for (DefaultWeightedEdge e : graph.edgeSet()) {
Integer alphaOcc = alphaWellCounts.get(plateVtoAMap.get(graph.getEdgeSource(e)));
@@ -66,20 +80,27 @@ public abstract class GraphModificationFunctions {
double weight = graph.getEdgeWeight(e);
double min = minOverlapPercent / 100.0;
if ((weight / alphaOcc < min) || (weight / betaOcc < min)) {
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer intWeight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, intWeight};
removedEdges.add(edge);
if(saveEdges) {
Integer source = graph.getEdgeSource(e);
Integer target = graph.getEdgeTarget(e);
Integer intWeight = (int) graph.getEdgeWeight(e);
Integer[] edge = {source, target, intWeight};
removedEdges.add(edge);
}
else {
graph.setEdgeWeight(e, 0.0);
}
}
}
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
if(saveEdges) {
for (Integer[] edge : removedEdges) {
graph.removeEdge(edge[0], edge[1]);
}
}
return removedEdges;
}
public static void addRemovedEdges(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
static void addRemovedEdges(SimpleWeightedGraph<Integer, DefaultWeightedEdge> graph,
List<Integer[]> removedEdges) {
for (Integer[] edge : removedEdges) {
DefaultWeightedEdge e = graph.addEdge(edge[0], edge[1]);

View File

@@ -11,7 +11,7 @@ public class GraphWithMapData implements java.io.Serializable {
private String sourceFilename;
private final SimpleWeightedGraph graph;
private Integer numWells;
private Integer[] wellConcentrations;
private Integer[] wellPopulations;
private Integer alphaCount;
private Integer betaCount;
private final Map<Integer, Integer> distCellsMapAlphaKey;
@@ -31,7 +31,7 @@ public class GraphWithMapData implements java.io.Serializable {
Map<Integer, Integer> betaWellCounts, Duration time) {
this.graph = graph;
this.numWells = numWells;
this.wellConcentrations = wellConcentrations;
this.wellPopulations = wellConcentrations;
this.alphaCount = alphaCount;
this.betaCount = betaCount;
this.distCellsMapAlphaKey = distCellsMapAlphaKey;
@@ -52,8 +52,8 @@ public class GraphWithMapData implements java.io.Serializable {
return numWells;
}
public Integer[] getWellConcentrations() {
return wellConcentrations;
public Integer[] getWellPopulations() {
return wellPopulations;
}
public Integer getAlphaCount() {

View File

@@ -1,14 +1,15 @@
import java.io.IOException;
import java.util.List;
import java.util.Scanner;
import java.util.InputMismatchException;
import java.util.*;
import java.util.regex.Matcher;
import java.util.regex.Pattern;
//
public class InteractiveInterface {
final static Scanner sc = new Scanner(System.in);
static int input;
static boolean quit = false;
private static final Random rand = BiGpairSEQ.getRand();
private static final Scanner sc = new Scanner(System.in);
private static int input;
private static boolean quit = false;
public static void startInteractive() {
@@ -26,6 +27,7 @@ public class InteractiveInterface {
//Need to re-do the CDR3/CDR1 matching to correspond to new pattern
//System.out.println("5) Generate CDR3/CDR1 occupancy graph");
//System.out.println("6) Simulate CDR3/CDR1 T cell matching");
System.out.println("8) Options");
System.out.println("9) About/Acknowledgments");
System.out.println("0) Exit");
try {
@@ -36,9 +38,10 @@ public class InteractiveInterface {
case 3 -> makeCDR3Graph();
case 4 -> matchCDR3s();
//case 6 -> matchCellsCDR1();
case 8 -> mainOptions();
case 9 -> acknowledge();
case 0 -> quit = true;
default -> throw new InputMismatchException("Invalid input.");
default -> System.out.println("Invalid input.");
}
} catch (InputMismatchException | IOException ex) {
System.out.println(ex);
@@ -71,10 +74,15 @@ public class InteractiveInterface {
System.out.println(ex);
sc.next();
}
CellSample sample = Simulator.generateCellSample(numCells, cdr1Freq);
CellSample sample = new CellSample(numCells, cdr1Freq);
assert filename != null;
System.out.println("Writing cells to file");
CellFileWriter writer = new CellFileWriter(filename, sample);
writer.writeCellsToFile();
System.out.println("Cell sample written to: " + filename);
if(BiGpairSEQ.cacheCells()) {
BiGpairSEQ.setCellSampleInMemory(sample, filename);
}
}
//Output a CSV of sample plate
@@ -84,7 +92,7 @@ public class InteractiveInterface {
Double stdDev = 0.0;
Integer numWells = 0;
Integer numSections;
Integer[] concentrations = {1};
Integer[] populations = {1};
Double dropOutRate = 0.0;
boolean poisson = false;
boolean exponential = false;
@@ -123,10 +131,11 @@ public class InteractiveInterface {
}
case 3 -> {
exponential = true;
System.out.println("Please enter lambda value for exponential distribution.");
System.out.print("Please enter lambda value for exponential distribution: ");
lambda = sc.nextDouble();
if (lambda <= 0.0) {
throw new InputMismatchException("Value must be positive.");
lambda = 0.6;
System.out.println("Value must be positive. Defaulting to 0.6.");
}
}
default -> {
@@ -139,22 +148,57 @@ public class InteractiveInterface {
if(numWells < 1){
throw new InputMismatchException("No wells on plate");
}
System.out.println("\nThe plate can be evenly sectioned to allow multiple concentrations of T-cells/well");
System.out.println("How many sections would you like to make (minimum 1)?");
numSections = sc.nextInt();
if(numSections < 1) {
throw new InputMismatchException("Too few sections.");
//choose whether to make T cell population/well random
boolean randomWellPopulations;
System.out.println("Randomize number of T cells in each well? (y/n)");
String ans = sc.next();
Pattern pattern = Pattern.compile("(?:yes|y)", Pattern.CASE_INSENSITIVE);
Matcher matcher = pattern.matcher(ans);
if(matcher.matches()){
randomWellPopulations = true;
}
else if (numSections > numWells) {
throw new InputMismatchException("Cannot have more sections than wells.");
else{
randomWellPopulations = false;
}
int i = 1;
concentrations = new Integer[numSections];
while(numSections > 0) {
System.out.print("Enter number of T-cells per well in section " + i +": ");
concentrations[i - 1] = sc.nextInt();
i++;
numSections--;
if(randomWellPopulations) { //if T cell population/well is random
numSections = numWells;
Integer minPop;
Integer maxPop;
System.out.print("Please enter minimum number of T cells in a well: ");
minPop = sc.nextInt();
if(minPop < 1) {
throw new InputMismatchException("Minimum well population must be positive");
}
System.out.println("Please enter maximum number of T cells in a well: ");
maxPop = sc.nextInt();
if(maxPop < minPop) {
throw new InputMismatchException("Max well population must be greater than min well population");
}
//maximum should be inclusive, so need to add one to max of randomly generated values
populations = rand.ints(minPop, maxPop + 1)
.limit(numSections)
.boxed()
.toArray(Integer[]::new);
System.out.print("Populations: ");
System.out.println(Arrays.toString(populations));
}
else{ //if T cell population/well is not random
System.out.println("\nThe plate can be evenly sectioned to allow different numbers of T cells per well.");
System.out.println("How many sections would you like to make (minimum 1)?");
numSections = sc.nextInt();
if (numSections < 1) {
throw new InputMismatchException("Too few sections.");
} else if (numSections > numWells) {
throw new InputMismatchException("Cannot have more sections than wells.");
}
int i = 1;
populations = new Integer[numSections];
while (numSections > 0) {
System.out.print("Enter number of T cells per well in section " + i + ": ");
populations[i - 1] = sc.nextInt();
i++;
numSections--;
}
}
System.out.println("\nErrors in amplification can induce a well dropout rate for sequences");
System.out.print("Enter well dropout rate (0.0 to 1.0): ");
@@ -166,26 +210,38 @@ public class InteractiveInterface {
System.out.println(ex);
sc.next();
}
System.out.println("Reading Cell Sample file: " + cellFile);
assert cellFile != null;
CellFileReader cellReader = new CellFileReader(cellFile);
CellSample cells;
if (cellFile.equals(BiGpairSEQ.getCellFilename())){
cells = BiGpairSEQ.getCellSampleInMemory();
}
else {
System.out.println("Reading Cell Sample file: " + cellFile);
CellFileReader cellReader = new CellFileReader(cellFile);
cells = cellReader.getCellSample();
if(BiGpairSEQ.cacheCells()) {
BiGpairSEQ.setCellSampleInMemory(cells, cellFile);
}
}
assert filename != null;
Plate samplePlate;
PlateFileWriter writer;
if(exponential){
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWellsExponential(cellReader.getFilename(), cellReader.getCells(), lambda);
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
writer.writePlateFile();
samplePlate = new Plate(cells, cellFile, numWells, populations, dropOutRate, lambda, true);
writer = new PlateFileWriter(filename, samplePlate);
}
else {
if (poisson) {
stdDev = Math.sqrt(cellReader.getCellCount()); //gaussian with square root of elements approximates poisson
stdDev = Math.sqrt(cells.getCellCount()); //gaussian with square root of elements approximates poisson
}
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWells(cellReader.getFilename(), cellReader.getCells(), stdDev);
assert filename != null;
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
System.out.println("Writing Sample Plate to file");
writer.writePlateFile();
System.out.println("Sample Plate written to file: " + filename);
samplePlate = new Plate(cells, cellFile, numWells, populations, dropOutRate, stdDev, false);
writer = new PlateFileWriter(filename, samplePlate);
}
System.out.println("Writing Sample Plate to file");
writer.writePlateFile();
System.out.println("Sample Plate written to file: " + filename);
if(BiGpairSEQ.cachePlate()) {
BiGpairSEQ.setPlateInMemory(samplePlate, filename);
}
}
@@ -194,7 +250,6 @@ public class InteractiveInterface {
String filename = null;
String cellFile = null;
String plateFile = null;
try {
String str = "\nGenerating bipartite weighted graph encoding occupancy overlap data ";
str = str.concat("\nrequires a cell sample file and a sample plate file.");
@@ -210,14 +265,37 @@ public class InteractiveInterface {
System.out.println(ex);
sc.next();
}
System.out.println("Reading Cell Sample file: " + cellFile);
assert cellFile != null;
CellFileReader cellReader = new CellFileReader(cellFile);
System.out.println("Reading Sample Plate file: " + plateFile);
CellSample cellSample;
//check if cells are already in memory
if(cellFile.equals(BiGpairSEQ.getCellFilename()) && BiGpairSEQ.getCellSampleInMemory() != null) {
cellSample = BiGpairSEQ.getCellSampleInMemory();
}
else {
System.out.println("Reading Cell Sample file: " + cellFile);
CellFileReader cellReader = new CellFileReader(cellFile);
cellSample = cellReader.getCellSample();
if(BiGpairSEQ.cacheCells()) {
BiGpairSEQ.setCellSampleInMemory(cellSample, cellFile);
}
}
assert plateFile != null;
PlateFileReader plateReader = new PlateFileReader(plateFile);
Plate plate = new Plate(plateReader.getFilename(), plateReader.getWells());
if (cellReader.getCells().size() == 0){
Plate plate;
//check if plate is already in memory
if(plateFile.equals(BiGpairSEQ.getPlateFilename())){
plate = BiGpairSEQ.getPlateInMemory();
}
else {
System.out.println("Reading Sample Plate file: " + plateFile);
PlateFileReader plateReader = new PlateFileReader(plateFile);
plate = plateReader.getSamplePlate();
if(BiGpairSEQ.cachePlate()) {
BiGpairSEQ.setPlateInMemory(plate, plateFile);
}
}
if (cellSample.getCells().size() == 0){
System.out.println("No cell sample found.");
System.out.println("Returning to main menu.");
}
@@ -226,15 +304,22 @@ public class InteractiveInterface {
System.out.println("Returning to main menu.");
}
else{
List<Integer[]> cells = cellReader.getCells();
GraphWithMapData data = Simulator.makeGraph(cells, plate, true);
GraphWithMapData data = Simulator.makeGraph(cellSample, plate, true);
assert filename != null;
GraphDataObjectWriter dataWriter = new GraphDataObjectWriter(filename, data);
dataWriter.writeDataToFile();
System.out.println("Graph and Data file written to: " + filename);
BiGpairSEQ.setGraph(data);
BiGpairSEQ.setGraphFilename(filename);
System.out.println("Graph and Data file " + filename + " cached.");
if(BiGpairSEQ.outputBinary()) {
GraphDataObjectWriter dataWriter = new GraphDataObjectWriter(filename, data);
dataWriter.writeDataToFile();
System.out.println("Serialized binary graph/data file written to: " + filename);
}
if(BiGpairSEQ.outputGraphML()) {
GraphMLFileWriter graphMLWriter = new GraphMLFileWriter(filename, data);
graphMLWriter.writeGraphToFile();
System.out.println("GraphML file written to: " + filename);
}
if(BiGpairSEQ.cacheGraph()) {
BiGpairSEQ.setGraphInMemory(data, filename);
}
}
}
@@ -256,17 +341,28 @@ public class InteractiveInterface {
System.out.println("\nWhat is the minimum number of CDR3 alpha/beta overlap wells to attempt matching?");
lowThreshold = sc.nextInt();
if(lowThreshold < 1){
throw new InputMismatchException("Minimum value for low threshold set to 1");
lowThreshold = 1;
System.out.println("Value for low occupancy overlap threshold must be positive");
System.out.println("Value for low occupancy overlap threshold set to 1");
}
System.out.println("\nWhat is the maximum number of CDR3 alpha/beta overlap wells to attempt matching?");
highThreshold = sc.nextInt();
System.out.println("\nWhat is the maximum difference in alpha/beta occupancy to attempt matching?");
maxOccupancyDiff = sc.nextInt();
System.out.println("\nWell overlap percentage = pair overlap / sequence occupancy");
System.out.println("What is the minimum well overlap percentage to attempt matching? (0 to 100)");
if(highThreshold < lowThreshold) {
highThreshold = lowThreshold;
System.out.println("Value for high occupancy overlap threshold must be >= low overlap threshold");
System.out.println("Value for high occupancy overlap threshold set to " + lowThreshold);
}
System.out.println("What is the minimum percentage of a sequence's wells in alpha/beta overlap to attempt matching? (0 - 100)");
minOverlapPercent = sc.nextInt();
if (minOverlapPercent < 0 || minOverlapPercent > 100) {
throw new InputMismatchException("Value outside range. Minimum percent set to 0");
System.out.println("Value outside range. Minimum occupancy overlap percentage set to 0");
}
System.out.println("\nWhat is the maximum difference in alpha/beta occupancy to attempt matching?");
maxOccupancyDiff = sc.nextInt();
if (maxOccupancyDiff < 0) {
maxOccupancyDiff = 0;
System.out.println("Maximum allowable difference in alpha/beta occupancy must be nonnegative");
System.out.println("Maximum allowable difference in alpha/beta occupancy set to 0");
}
} catch (InputMismatchException ex) {
System.out.println(ex);
@@ -275,17 +371,15 @@ public class InteractiveInterface {
assert graphFilename != null;
//check if this is the same graph we already have in memory.
GraphWithMapData data;
if(!(graphFilename.equals(BiGpairSEQ.getGraphFilename())) || BiGpairSEQ.getGraph() == null) {
BiGpairSEQ.clearGraph();
//read object data from file
GraphDataObjectReader dataReader = new GraphDataObjectReader(graphFilename);
data = dataReader.getData();
//set new graph in memory and new filename
BiGpairSEQ.setGraph(data);
BiGpairSEQ.setGraphFilename(graphFilename);
if(graphFilename.equals(BiGpairSEQ.getGraphFilename())) {
data = BiGpairSEQ.getGraphInMemory();
}
else {
data = BiGpairSEQ.getGraph();
GraphDataObjectReader dataReader = new GraphDataObjectReader(graphFilename, true);
data = dataReader.getData();
if(BiGpairSEQ.cacheGraph()) {
BiGpairSEQ.setGraphInMemory(data, graphFilename);
}
}
//simulate matching
MatchingResult results = Simulator.matchCDR3s(data, graphFilename, lowThreshold, highThreshold, maxOccupancyDiff,
@@ -402,6 +496,79 @@ public class InteractiveInterface {
// }
// }
private static void mainOptions(){
boolean backToMain = false;
while(!backToMain) {
System.out.println("\n--------------OPTIONS---------------");
System.out.println("1) Turn " + getOnOff(!BiGpairSEQ.cacheCells()) + " cell sample file caching");
System.out.println("2) Turn " + getOnOff(!BiGpairSEQ.cachePlate()) + " plate file caching");
System.out.println("3) Turn " + getOnOff(!BiGpairSEQ.cacheGraph()) + " graph/data file caching");
System.out.println("4) Turn " + getOnOff(!BiGpairSEQ.outputBinary()) + " serialized binary graph output");
System.out.println("5) Turn " + getOnOff(!BiGpairSEQ.outputGraphML()) + " GraphML graph output");
System.out.println("6) Maximum weight matching algorithm options");
System.out.println("0) Return to main menu");
try {
input = sc.nextInt();
switch (input) {
case 1 -> BiGpairSEQ.setCacheCells(!BiGpairSEQ.cacheCells());
case 2 -> BiGpairSEQ.setCachePlate(!BiGpairSEQ.cachePlate());
case 3 -> BiGpairSEQ.setCacheGraph(!BiGpairSEQ.cacheGraph());
case 4 -> BiGpairSEQ.setOutputBinary(!BiGpairSEQ.outputBinary());
case 5 -> BiGpairSEQ.setOutputGraphML(!BiGpairSEQ.outputGraphML());
case 6 -> algorithmOptions();
case 0 -> backToMain = true;
default -> System.out.println("Invalid input");
}
} catch (InputMismatchException ex) {
System.out.println(ex);
sc.next();
}
}
}
/**
* Helper function for printing menu items in mainOptions(). Returns a string based on the value of parameter.
*
* @param b - a boolean value
* @return String "on" if b is true, "off" if b is false
*/
private static String getOnOff(boolean b) {
if (b) { return "on";}
else { return "off"; }
}
private static void algorithmOptions(){
boolean backToOptions = false;
while(!backToOptions) {
System.out.println("\n---------ALGORITHM OPTIONS----------");
System.out.println("1) Use scaling algorithm by Duan and Su.");
System.out.println("2) Use LEDA book algorithm with Fibonacci heap priority queue");
System.out.println("3) Use LEDA book algorithm with pairing heap priority queue");
System.out.println("0) Return to Options menu");
try {
input = sc.nextInt();
switch (input) {
case 1 -> System.out.println("This option is not yet implemented. Choose another.");
case 2 -> {
BiGpairSEQ.setFibonacciHeap();
System.out.println("MWM algorithm set to LEDA with Fibonacci heap");
backToOptions = true;
}
case 3 -> {
BiGpairSEQ.setPairingHeap();
System.out.println("MWM algorithm set to LEDA with pairing heap");
backToOptions = true;
}
case 0 -> backToOptions = true;
default -> System.out.println("Invalid input");
}
} catch (InputMismatchException ex) {
System.out.println(ex);
sc.next();
}
}
}
private static void acknowledge(){
System.out.println("This program simulates BiGpairSEQ, a graph theory based adaptation");
System.out.println("of the pairSEQ algorithm for pairing T cell receptor sequences.");

View File

@@ -21,15 +21,15 @@ public class MatchingResult {
* well populations *
* total alphas found *
* total betas found *
* high overlap threshold
* low overlap threshold
* maximum occupancy difference
* minimum overlap percent
* pairing attempt rate
* correct pairing count
* incorrect pairing count
* pairing error rate
* simulation time
* high overlap threshold *
* low overlap threshold *
* maximum occupancy difference *
* minimum overlap percent *
* pairing attempt rate *
* correct pairing count *
* incorrect pairing count *
* pairing error rate *
* simulation time (seconds)
*/
this.metadata = metadata;
this.comments = new ArrayList<>();
@@ -91,6 +91,22 @@ public class MatchingResult {
return Integer.parseInt(metadata.get("total beta count"));
}
//put in the rest of these methods following the same pattern
public Integer getHighOverlapThreshold() { return Integer.parseInt(metadata.get("high overlap threshold"));}
public Integer getLowOverlapThreshold() { return Integer.parseInt(metadata.get("low overlap threshold"));}
public Integer getMaxOccupancyDifference() { return Integer.parseInt(metadata.get("maximum occupancy difference"));}
public Integer getMinOverlapPercent() { return Integer.parseInt(metadata.get("minimum overlap percent"));}
public Double getPairingAttemptRate() { return Double.parseDouble(metadata.get("pairing attempt rate"));}
public Integer getCorrectPairingCount() { return Integer.parseInt(metadata.get("correct pairing count"));}
public Integer getIncorrectPairingCount() { return Integer.parseInt(metadata.get("incorrect pairing count"));}
public Double getPairingErrorRate() { return Double.parseDouble(metadata.get("pairing error rate"));}
public String getSimulationTime() { return metadata.get("simulation time (seconds)"); }
}

View File

@@ -8,9 +8,11 @@ TODO: Implement discrete frequency distributions using Vose's Alias Method
import java.util.*;
public class Plate {
private CellSample cells;
private String sourceFile;
private String filename;
private List<List<Integer[]>> wells;
private Random rand = new Random();
private final Random rand = BiGpairSEQ.getRand();
private int size;
private double error;
private Integer[] populations;
@@ -18,6 +20,25 @@ public class Plate {
private double lambda;
boolean exponential = false;
public Plate(CellSample cells, String cellFilename, int numWells, Integer[] populations,
double dropoutRate, double stdDev_or_lambda, boolean exponential){
this.cells = cells;
this.sourceFile = cellFilename;
this.size = numWells;
this.wells = new ArrayList<>();
this.error = dropoutRate;
this.populations = populations;
this.exponential = exponential;
if (this.exponential) {
this.lambda = stdDev_or_lambda;
fillWellsExponential(cells.getCells(), this.lambda);
}
else {
this.stdDev = stdDev_or_lambda;
fillWells(cells.getCells(), this.stdDev);
}
}
public Plate(int size, double error, Integer[] populations) {
this.size = size;
@@ -26,8 +47,9 @@ public class Plate {
wells = new ArrayList<>();
}
public Plate(String sourceFileName, List<List<Integer[]>> wells) {
this.sourceFile = sourceFileName;
//constructor for returning a Plate from a PlateFileReader
public Plate(String filename, List<List<Integer[]>> wells) {
this.filename = filename;
this.wells = wells;
this.size = wells.size();
@@ -43,15 +65,13 @@ public class Plate {
}
}
public void fillWellsExponential(String sourceFileName, List<Integer[]> cells, double lambda){
private void fillWellsExponential(List<Integer[]> cells, double lambda){
this.lambda = lambda;
exponential = true;
sourceFile = sourceFileName;
int numSections = populations.length;
int section = 0;
double m;
int n;
int test=0;
while (section < numSections){
for (int i = 0; i < (size / numSections); i++) {
List<Integer[]> well = new ArrayList<>();
@@ -61,13 +81,6 @@ public class Plate {
m = (Math.log10((1 - rand.nextDouble()))/(-lambda)) * Math.sqrt(cells.size());
} while (m >= cells.size() || m < 0);
n = (int) Math.floor(m);
//n = Equations.getRandomNumber(0, cells.size());
// was testing generating the cell sample file with exponential dist, then sampling flat here
//that would be more realistic
//But would mess up other things in the simulation with how I've coded it.
if(n > test){
test = n;
}
Integer[] cellToAdd = cells.get(n).clone();
for(int k = 0; k < cellToAdd.length; k++){
if(Math.abs(rand.nextDouble()) < error){//error applied to each seqeunce
@@ -80,12 +93,10 @@ public class Plate {
}
section++;
}
System.out.println("Highest index: " +test);
}
public void fillWells(String sourceFileName, List<Integer[]> cells, double stdDev) {
private void fillWells( List<Integer[]> cells, double stdDev) {
this.stdDev = stdDev;
sourceFile = sourceFileName;
int numSections = populations.length;
int section = 0;
double m;
@@ -168,4 +179,6 @@ public class Plate {
public String getSourceFileName() {
return sourceFile;
}
public String getFilename() { return filename; }
}

View File

@@ -56,11 +56,8 @@ public class PlateFileReader {
}
public List<List<Integer[]>> getWells() {
return wells;
public Plate getSamplePlate() {
return new Plate(filename, wells);
}
public String getFilename() {
return filename;
}
}

View File

@@ -16,7 +16,7 @@ public class PlateFileWriter {
private Double error;
private String filename;
private String sourceFileName;
private Integer[] concentrations;
private Integer[] populations;
private boolean isExponential = false;
public PlateFileWriter(String filename, Plate plate) {
@@ -35,8 +35,8 @@ public class PlateFileWriter {
}
this.error = plate.getError();
this.wells = plate.getWells();
this.concentrations = plate.getPopulations();
Arrays.sort(concentrations);
this.populations = plate.getPopulations();
Arrays.sort(populations);
}
public void writePlateFile(){
@@ -73,14 +73,12 @@ public class PlateFileWriter {
// rows.add(tmp);
// }
//get list of well populations
List<Integer> wellPopulations = Arrays.asList(concentrations);
//make string out of populations list
//make string out of populations array
StringBuilder populationsStringBuilder = new StringBuilder();
populationsStringBuilder.append(wellPopulations.remove(0).toString());
for(Integer i: wellPopulations){
populationsStringBuilder.append(populations[0].toString());
for(int i = 1; i < populations.length; i++){
populationsStringBuilder.append(", ");
populationsStringBuilder.append(i.toString());
populationsStringBuilder.append(populations[i].toString());
}
String wellPopulationsString = populationsStringBuilder.toString();

View File

@@ -1,9 +1,9 @@
import org.jgrapht.Graph;
import org.jgrapht.alg.interfaces.MatchingAlgorithm;
import org.jgrapht.alg.matching.MaximumWeightBipartiteMatching;
import org.jgrapht.generate.SimpleWeightedBipartiteGraphMatrixGenerator;
import org.jgrapht.graph.DefaultWeightedEdge;
import org.jgrapht.graph.SimpleWeightedGraph;
import org.jheaps.tree.FibonacciHeap;
import org.jheaps.tree.PairingHeap;
import java.math.BigDecimal;
@@ -14,40 +14,19 @@ import java.time.Duration;
import java.util.*;
import java.util.stream.IntStream;
import static java.lang.Float.*;
//NOTE: "sequence" in method and variable names refers to a peptide sequence from a simulated T cell
public class Simulator {
public class Simulator implements GraphModificationFunctions {
private static final int cdr3AlphaIndex = 0;
private static final int cdr3BetaIndex = 1;
private static final int cdr1AlphaIndex = 2;
private static final int cdr1BetaIndex = 3;
public static CellSample generateCellSample(Integer numDistinctCells, Integer cdr1Freq) {
//In real T cells, CDR1s have about one third the diversity of CDR3s
List<Integer> numbersCDR3 = new ArrayList<>();
List<Integer> numbersCDR1 = new ArrayList<>();
Integer numDistCDR3s = 2 * numDistinctCells + 1;
IntStream.range(1, numDistCDR3s + 1).forEach(i -> numbersCDR3.add(i));
IntStream.range(numDistCDR3s + 1, numDistCDR3s + 1 + (numDistCDR3s / cdr1Freq) + 1).forEach(i -> numbersCDR1.add(i));
Collections.shuffle(numbersCDR3);
Collections.shuffle(numbersCDR1);
//Each cell represented by 4 values
//two CDR3s, and two CDR1s. First two values are CDR3s (alpha, beta), second two are CDR1s (alpha, beta)
List<Integer[]> distinctCells = new ArrayList<>();
for(int i = 0; i < numbersCDR3.size() - 1; i = i + 2){
Integer tmpCDR3a = numbersCDR3.get(i);
Integer tmpCDR3b = numbersCDR3.get(i+1);
Integer tmpCDR1a = numbersCDR1.get(i % numbersCDR1.size());
Integer tmpCDR1b = numbersCDR1.get((i+1) % numbersCDR1.size());
Integer[] tmp = {tmpCDR3a, tmpCDR3b, tmpCDR1a, tmpCDR1b};
distinctCells.add(tmp);
}
return new CellSample(distinctCells, cdr1Freq);
}
//Make the graph needed for matching CDR3s
public static GraphWithMapData makeGraph(List<Integer[]> distinctCells, Plate samplePlate, boolean verbose) {
public static GraphWithMapData makeGraph(CellSample cellSample, Plate samplePlate, boolean verbose) {
Instant start = Instant.now();
List<Integer[]> distinctCells = cellSample.getCells();
int[] alphaIndex = {cdr3AlphaIndex};
int[] betaIndex = {cdr3BetaIndex};
@@ -135,7 +114,7 @@ public class Simulator {
distCellsMapAlphaKey, plateVtoAMap, plateVtoBMap, plateAtoVMap,
plateBtoVMap, alphaWellCounts, betaWellCounts, time);
//Set source file name in graph to name of sample plate
output.setSourceFilename(samplePlate.getSourceFileName());
output.setSourceFilename(samplePlate.getFilename());
//return GraphWithMapData object
return output;
}
@@ -145,8 +124,8 @@ public class Simulator {
Integer highThreshold, Integer maxOccupancyDifference,
Integer minOverlapPercent, boolean verbose) {
Instant start = Instant.now();
//Integer arrays will contain TO VERTEX, FROM VERTEX, and WEIGHT (which I'll need to cast to double)
List<Integer[]> removedEdges = new ArrayList<>();
boolean saveEdges = BiGpairSEQ.cacheGraph();
int numWells = data.getNumWells();
Integer alphaCount = data.getAlphaCount();
Integer betaCount = data.getBetaCount();
@@ -159,33 +138,50 @@ public class Simulator {
//remove edges with weights outside given overlap thresholds, add those to removed edge list
if(verbose){System.out.println("Eliminating edges with weights outside overlap threshold values");}
removedEdges.addAll(GraphModificationFunctions.filterByOverlapThresholds(graph, lowThreshold, highThreshold));
removedEdges.addAll(GraphModificationFunctions.filterByOverlapThresholds(graph, lowThreshold, highThreshold, saveEdges));
if(verbose){System.out.println("Over- and under-weight edges removed");}
//remove edges between vertices with too small an overlap size, add those to removed edge list
if(verbose){System.out.println("Eliminating edges with weights less than " + minOverlapPercent.toString() +
" percent of vertex occupancy value.");}
removedEdges.addAll(GraphModificationFunctions.filterByOverlapPercent(graph, alphaWellCounts, betaWellCounts,
plateVtoAMap, plateVtoBMap, minOverlapPercent));
plateVtoAMap, plateVtoBMap, minOverlapPercent, saveEdges));
if(verbose){System.out.println("Edges with weights too far below a vertex occupancy value removed");}
//Filter by relative occupancy
if(verbose){System.out.println("Eliminating edges between vertices with occupancy difference > "
+ maxOccupancyDifference);}
removedEdges.addAll(GraphModificationFunctions.filterByRelativeOccupancy(graph, alphaWellCounts, betaWellCounts,
plateVtoAMap, plateVtoBMap, maxOccupancyDifference));
plateVtoAMap, plateVtoBMap, maxOccupancyDifference, saveEdges));
if(verbose){System.out.println("Edges between vertices of with excessively different occupancy values " +
"removed");}
//Find Maximum Weighted Matching
//using jheaps library class PairingHeap for improved efficiency
if(verbose){System.out.println("Finding maximum weighted matching");}
//Attempting to use addressable heap to improve performance
MaximumWeightBipartiteMatching maxWeightMatching =
new MaximumWeightBipartiteMatching(graph,
MaximumWeightBipartiteMatching maxWeightMatching;
//Use correct heap type for priority queue
String heapType = BiGpairSEQ.getPriorityQueueHeapType();
switch (heapType) {
case "PAIRING" -> {
maxWeightMatching = new MaximumWeightBipartiteMatching(graph,
plateVtoAMap.keySet(),
plateVtoBMap.keySet(),
i -> new PairingHeap(Comparator.naturalOrder()));
}
case "FIBONACCI" -> {
maxWeightMatching = new MaximumWeightBipartiteMatching(graph,
plateVtoAMap.keySet(),
plateVtoBMap.keySet(),
i -> new FibonacciHeap(Comparator.naturalOrder()));
}
default -> {
maxWeightMatching = new MaximumWeightBipartiteMatching(graph,
plateVtoAMap.keySet(),
plateVtoBMap.keySet());
}
}
//get the matching
MatchingAlgorithm.Matching<String, DefaultWeightedEdge> graphMatching = maxWeightMatching.getMatching();
if(verbose){System.out.println("Matching completed");}
Instant stop = Instant.now();
@@ -241,16 +237,23 @@ public class Simulator {
}
//Metadata comments for CSV file
String algoType = "LEDA book with heap: " + heapType;
int min = Math.min(alphaCount, betaCount);
//rate of attempted matching
double attemptRate = (double) (trueCount + falseCount) / min;
BigDecimal attemptRateTrunc = new BigDecimal(attemptRate, mc);
//rate of pairing error
double pairingErrorRate = (double) falseCount / (trueCount + falseCount);
BigDecimal pairingErrorRateTrunc = new BigDecimal(pairingErrorRate, mc);
//get list of well concentrations
Integer[] wellPopulations = data.getWellConcentrations();
//make string out of concentrations list
BigDecimal pairingErrorRateTrunc;
if(pairingErrorRate == NaN || pairingErrorRate == POSITIVE_INFINITY || pairingErrorRate == NEGATIVE_INFINITY) {
pairingErrorRateTrunc = new BigDecimal(-1, mc);
}
else{
pairingErrorRateTrunc = new BigDecimal(pairingErrorRate, mc);
}
//get list of well populations
Integer[] wellPopulations = data.getWellPopulations();
//make string out of populations list
StringBuilder populationsStringBuilder = new StringBuilder();
populationsStringBuilder.append(wellPopulations[0].toString());
for(int i = 1; i < wellPopulations.length; i++){
@@ -265,18 +268,19 @@ public class Simulator {
Map<String, String> metadata = new LinkedHashMap<>();
metadata.put("sample plate filename", data.getSourceFilename());
metadata.put("graph filename", dataFilename);
metadata.put("algorithm type", algoType);
metadata.put("well populations", wellPopulationsString);
metadata.put("total alphas found", alphaCount.toString());
metadata.put("total betas found", betaCount.toString());
metadata.put("high overlap threshold", highThreshold.toString());
metadata.put("low overlap threshold", lowThreshold.toString());
metadata.put("maximum occupancy difference", maxOccupancyDifference.toString());
metadata.put("minimum overlap percent", minOverlapPercent.toString());
metadata.put("maximum occupancy difference", maxOccupancyDifference.toString());
metadata.put("pairing attempt rate", attemptRateTrunc.toString());
metadata.put("correct pairing count", Integer.toString(trueCount));
metadata.put("incorrect pairing count", Integer.toString(falseCount));
metadata.put("pairing error rate", pairingErrorRateTrunc.toString());
metadata.put("simulation time", nf.format(time.toSeconds()));
metadata.put("simulation time (seconds)", nf.format(time.toSeconds()));
//create MatchingResult object
MatchingResult output = new MatchingResult(metadata, header, allResults, matchMap, time);
if(verbose){
@@ -285,10 +289,11 @@ public class Simulator {
}
}
//put the removed edges back on the graph
System.out.println("Restoring removed edges to graph.");
GraphModificationFunctions.addRemovedEdges(graph, removedEdges);
if(saveEdges) {
//put the removed edges back on the graph
System.out.println("Restoring removed edges to graph.");
GraphModificationFunctions.addRemovedEdges(graph, removedEdges);
}
//return MatchingResult object
return output;
}
@@ -664,7 +669,7 @@ public class Simulator {
private static Map<Integer, Integer> makeVertexToSequenceMap(Map<Integer, Integer> sequences, Integer startValue) {
Map<Integer, Integer> map = new LinkedHashMap<>(); //LinkedHashMap to preserve order of entry
Integer index = startValue;
Integer index = startValue; //is this necessary? I don't think I use this.
for (Integer k: sequences.keySet()) {
map.put(index, k);
index++;

View File

@@ -1,14 +1,20 @@
public class Vertex {
private final Integer peptide;
private final Integer vertexLabel;
private final Integer sequence;
private final Integer occupancy;
public Vertex(Integer peptide, Integer occupancy) {
this.peptide = peptide;
public Vertex(Integer vertexLabel, Integer sequence, Integer occupancy) {
this.vertexLabel = vertexLabel;
this.sequence = sequence;
this.occupancy = occupancy;
}
public Integer getPeptide() {
return peptide;
public Integer getVertexLabel() { return vertexLabel; }
public Integer getSequence() {
return sequence;
}
public Integer getOccupancy() {