Implemented parameter for CDR1 frequency

src/main/java/CellFileWriter.java

@@ -13,6 +13,7 @@ public class CellFileWriter {
     private String[] headers = {"Alpha CDR3", "Beta CDR3", "Alpha CDR1", "Beta CDR1"};
     List<Integer[]> cells;
     String filename;
+    Integer cdr1Freq;
 
     public CellFileWriter(String filename, CellSample cells) {
         if(!filename.matches(".*\\.csv")){
@@ -20,15 +21,18 @@ public class CellFileWriter {
         }
         this.filename = filename;
         this.cells = cells.getCells();
+        this.cdr1Freq = cells.getCdr1Freq();
     }
 
     public void writeCellsToFile() {
         CSVFormat cellFileFormat = CSVFormat.Builder.create()
+                .setCommentMarker('#')
                 .setHeader(headers)
                 .build();
         try(BufferedWriter writer = Files.newBufferedWriter(Path.of(filename), StandardOpenOption.CREATE_NEW);
             CSVPrinter printer = new CSVPrinter(writer, cellFileFormat);
         ){
+            printer.printComment("Sample contains 1 unique CDR1 for every " + cdr1Freq + "unique CDR3s.");
             printer.printRecords(cells);
         } catch(IOException ex){
             System.out.println("Could not make new file named "+filename);

src/main/java/CellSample.java

@@ -3,15 +3,21 @@ import java.util.List;
 public class CellSample {
 
     private List<Integer[]> cells;
+    private Integer cdr1Freq;
 
-    public CellSample(List<Integer[]> cells){
+    public CellSample(List<Integer[]> cells, Integer cdr1Freq){
         this.cells = cells;
+        this.cdr1Freq = cdr1Freq;
     }
 
     public List<Integer[]> getCells(){
         return cells;
     }
 
+    public Integer getCdr1Freq() {
+        return cdr1Freq;
+    }
+
     public Integer population(){
         return cells.size();
     }

src/main/java/Simulator.java

@@ -29,31 +29,14 @@ public class Simulator {
     private static boolean useJGraphTGraphMatrixGenerator = true; //fastest option
     */
 
-    public static CellSample generateCellSample(Integer numDistinctCells) {
-        List<Integer> numbers = new ArrayList<>();
-        IntStream.range(1, (2 * numDistinctCells) + 1).forEach(i -> numbers.add(i));
-        Collections.shuffle(numbers);
-
-        //Each cell represented by two numbers from the random permutation
-        //These represent unique alpha and beta peptides
-        List<Integer[]> distinctCells = new ArrayList<>();
-        for(int i = 0; i < numbers.size() - 1; i = i + 2) {
-            Integer tmp1 = numbers.get(i);
-            Integer tmp2 = numbers.get(i+1);
-            Integer[] tmp = {tmp1, tmp2};
-            distinctCells.add(tmp);
-        }
-        return new CellSample(distinctCells);
-    }
-
-    public static CellSample generateExpandedCellSample(Integer numDistinctCells) {
+    public static CellSample generateCellSample(Integer numDistinctCells, Integer cdr1Freq) {
         //In real T cells, CDR1s have about one third the diversity of CDR3s
         //previous sim was only 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 / 3) + 1).forEach(i -> numbersCDR1.add(i));
+        IntStream.range(numDistCDR3s + 1, numDistCDR3s + 1 + (numDistCDR3s / cdr1Freq) + 1).forEach(i -> numbersCDR1.add(i));
         Collections.shuffle(numbersCDR3);
         Collections.shuffle(numbersCDR1);
 
@@ -68,7 +51,7 @@ public class Simulator {
             Integer[] tmp = {tmpCDR3a, tmpCDR3b, tmpCDR1a, tmpCDR1b};
             distinctCells.add(tmp);
         }
-        return new CellSample(distinctCells);
+        return new CellSample(distinctCells, cdr1Freq);
     }
 
     public static MatchingResult matchCDR3s(List<Integer[]> distinctCells,

src/main/java/UserInterface.java

@@ -41,6 +41,7 @@ public class UserInterface {
     private static void makeCells() {
         String filename = null;
         Integer numCells = 0;
+        Integer cdr1Freq = 1;
         try {
             System.out.println("\nSimulated T-Cells consist of integer values representing:\n" +
                     "* a pair of alpha and beta CDR3 peptides (unique within simulated population)\n" +
@@ -48,6 +49,9 @@ public class UserInterface {
             System.out.println("\nThe cells will be written to a file.");
             System.out.print("Please enter a file name: ");
             filename = sc.next();
+            System.out.println("CDR3 sequences are more diverse than CDR1 sequences.");
+            System.out.println("Please enter the factor by which distinct CDR3s outnumber CDR1s: ");
+            cdr1Freq = sc.nextInt();
             System.out.print("Please enter the number of T-cells to generate: ");
             numCells = sc.nextInt();
             if(numCells <= 0){
@@ -57,7 +61,7 @@ public class UserInterface {
             System.out.println(ex);
             sc.next();
         }
-        CellSample sample = Simulator.generateExpandedCellSample(numCells);
+        CellSample sample = Simulator.generateCellSample(numCells, cdr1Freq);
         CellFileWriter writer = new CellFileWriter(filename, sample);
         writer.writeCellsToFile();
     }