Files
BiGpairSEQ/src/main/java/CommandLineInterface.java
2022-02-23 10:30:42 -06:00

329 lines
15 KiB
Java

import org.apache.commons.cli.*;
/*
* Class for parsing options passed to program from command line
*
* Top-level flags:
* cells : to make a cell sample file
* plate : to make a sample plate file
* graph : to make a graph and data file
* match : to do a cdr3 matching (WITH OR WITHOUT MAKING A RESULTS FILE. May just want to print summary for piping.)
*
* Cell flags:
* count : number of cells to generate
* diversity factor : factor by which CDR3s are more diverse than CDR1s
* output : name of the output file
*
* Plate flags:
* cellfile : name of the cell sample file to use as input
* wells : the number of wells on the plate
* dist : the statistical distribution to use
* (if exponential) lambda : the lambda value of the exponential distribution
* (if gaussian) stddev : the standard deviation of the gaussian distribution
* rand : randomize well populations, take a minimum argument and a maximum argument
* populations : number of t cells per well per section (number of arguments determines number of sections)
* dropout : plate dropout rate, double from 0.0 to 1.0
* output : name of the output file
*
* Graph flags:
* 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
*
* Match flags:
* graphFile : name of graph and data file to use as input
* min : minimum number of overlap wells to attempt a matching
* max : the maximum number of overlap wells to attempt a matching
* maxdiff : (optional) the maximum difference in occupancy to attempt a matching
* minpercent : (optional) the minimum percent overlap to attempt a matching.
* writefile : (optional) the filename to write results to
* output : the values to print to System.out for piping
*
*/
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);
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);
}
}
else if(line.hasOption("cells")){
//line = parser.parse(mainOptions, args);
String filename = line.getOptionValue("o");
Integer numDistCells = Integer.valueOf(line.getOptionValue("nc"));
Integer freq = Integer.valueOf(line.getOptionValue("d"));
makeCells(filename, numDistCells, freq);
}
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]);
}
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("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 if(line.hasOption("poisson")){
for(int i = 1; i <= numPlatesToMake; i++){
makePlatePoisson(cellFile, filenamePrefix + i, numWellsOnPlate,
concentrationsToUse,dropOutRate);
}
}
}
}
catch (ParseException exp) {
System.err.println("Parsing failed. Reason: " + exp.getMessage());
}
}
//for calling from command line
public static void makeCells(String filename, Integer numCells, Integer cdr1Freq){
CellSample sample = Simulator.generateCellSample(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.getListOfDistinctCellsDEPRECATED(), 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.getCellCountDEPRECATED());
Plate samplePlate = new Plate(numWells, dropOutRate, concentrations);
samplePlate.fillWells(cellReader.getFilename(), cellReader.getListOfDistinctCellsDEPRECATED(), 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.getListOfDistinctCellsDEPRECATED(), stdDev);
PlateFileWriter writer = new PlateFileWriter(filename, samplePlate);
writer.writePlateFile();
}
private static void matchCDR3s(String graphFile, Integer lowThreshold, Integer highThreshold,
Integer occupancyDifference, Integer overlapPercent) {
}
}