Merge pull request #250 from bpt26/master

add simulation scripts

scripts/recombination/simulation/README.md

@@ -0,0 +1,42 @@
+# Recombination Simulation
+
+This repository contains scripts to create a multiple-sequence alignment of recombinant genomes. It was designed to test the recombination module in [UShER](https://usher-wiki.readthedocs.io/en/latest/), but may be used for other purposes as well.
+
+Note: I have been using specifically commit a6f65ade7a6606ef75902ee290585c6db23aeba6 as after this, the format for the output of `matUtils extract -S` has changed. To get this version of user, you can use `wget https://github.com/yatisht/usher/archive/a6f65ade7a6606ef75902ee290585c6db23aeba6.zip` and follow the remainder of the install instructions at [the UShER Wiki](https://usher-wiki.readthedocs.io/en/latest/).
+
+## Summary:
+
+### makeMutsFile.py
+generates a MSA of all nodes of interest. By default, this will produce a diff file for each internal node with at least 10 descendants, but you may set -t to 0 to get this for all internal nodes and leaves, or set -f to output a full MSA as well.
+
+the internal nodes from the initial tree with at least 10 descendants.  
+Options:
+- -s (--samples): Sample mutation paths file, generated by e.g. *matUtils extract -i <input.pb> -A <sample-paths.tsv>*.  
+- -l (--leaves): Leaves per node, generated by e.g. *matUtils extract -i <input.pb> -L <num-leaves.tsv>*.  
+- -t (--threshold): By default, this script includes only nodes with at least 10 descendant leaves. To use a different minimum, specify here. (Default = 10).  
+- -r (--reference): Reference genome, used to generate VCF. Default = 'wuhan.ref.fa' (in this repository, corresponds to [the reference genome of SARS-CoV-2](https://www.ncbi.nlm.nih.gov/nuccore/1798174254) with [problematic sites](https://raw.githubusercontent.com/W-L/ProblematicSites_SARS-CoV2/master/problematic_sites_sarsCov2.vcf) masked.).
+- -f (--fasta): Write fasta file containing all samples according to their mutation paths. This makes the script significantly slower. (Default = False)  
+
+### makeRandomRecombinants.py
+generates a MSA of recombinant samples from the internal node sequences.  
+Options:
+- -b (--breakpoints): Number of breakpoints that each recombinant sample will have. Must be 1 or 2 (Default = 1).  
+- -s (--samples): Number of recombinant samples to create (Default = 100).  
+- -c (--copies): Number of identical copies to make for each recombinant sample (Default = 10).  
+- -d (--differences): Minimum mutational distance for acceptor/donor samples (Default = 10).  
+- -m (--commonMutations): Number of mutations to add to each copy, shared by all in a set. (Default = 0).  
+- -M (--randomMutations): Number of mutations to add to each copy, randomly chosen for each copy. (Default = 0).  
+- -f (--fasta): Fasta file containing sequences for acceptor/donor samples. [REQUIRED]  
+- -r (--reference): Fasta file containing reference genome for use in creating VCF. (Default = 'wuhan.ref.fa').  
+- -S (--separate): If enabled, will produce one MSA as a .fasta file for each set of recombinants to the argument directory. If not enabled, will not produce these files.  
+
+## Example pipeline:
+
+```
+matUtils extract -i input.pb -S samples.tsv -L num_leaves.tsv
+python makeMutsFile.py -s samples.tsv -l num-leaves.txt -t 10 -r wuhan.ref.fa  
+python makeRandomRecombinants.py -b 1 -s 100 -c 10 -t 10 -d allNodeToMutsT10.txt -r wuhan.ref.fa -S 1BP_0M_10C
+faToVcf recombination_1.msa.fa recombination_1.msa.vcf
+usher -i input.pb -v recombination_1.msa.vcf -o recombination_1.msa.pb
+findRecombination -i recombination_1.msa.pb
+```

scripts/recombination/simulation/makeInternalNodesMSA.py

@@ -0,0 +1,200 @@
+#!/usr/bin/env python3
+# Name: Bryan Thornlow
+# Date: 2/1/2018
+# compareDatabases.py
+
+import sys
+import os
+import datetime
+import numpy
+from numpy import random
+import gzip
+import math
+import argparse
+
+##########################
+###### COMMAND LINE ######
+##########################
+
+class CommandLine(object):
+    """
+    Handles the input arguments from the command line. Manages 
+    the argument parser.
+    """
+
+    def __init__(self, inOpts=None):
+        '''
+        CommandLine constructor.
+        Implements a parser to interpret the command line input using argparse.
+        '''
+        self.parser = argparse.ArgumentParser()
+        self.parser.add_argument("-s", "--samples", help="Sample mutation paths file. To generate: matUtils extract -i <input.pb> -A <sample-paths.tsv>. [REQUIRED]", default='')
+        self.parser.add_argument("-l", "--leaves", help="File containing number of leaves for each node. Only nodes with at least 10 descendants will be used, "+
+            "as this is required for detection. To generate: matUtils extract -i <input.pb -L <num-leaves.tsv>. [REQUIRED]", default='')
+        self.parser.add_argument("-t", "--threshold", help="By default, this script includes only nodes with at least 10 descendant leaves. To use a different minimum, specify here.",default=10)
+        self.parser.add_argument("-r", "--ref", help="Fasta file containing reference genome. (Default = 'wuhan.ref.fa').", default='wuhan.ref.fa')
+
+        if inOpts is None:
+            self.args = vars(self.parser.parse_args())
+        else:
+            self.args = vars(self.parser.parse_args(inOpts))
+        self.args = self.parser.parse_args()     
+
+##########################
+##### MAIN FUNCTIONS #####
+##########################
+
+def getMutationsFile(myS, myL, myT, myR):
+    myReference = ''
+    with open(myR) as f:
+        for line in f:
+            l = line.strip()
+            if not l.startswith('>'):
+                myReference += l.strip().upper()
+
+    nodeToMuts = {1:[]}
+    with open(myS) as f:
+        for line in f:
+            splitLine = (line.strip()).split('\t')
+            myPath = stripEach(splitLine[1].split('>'))
+            pastMuts = []
+            for k in myPath:
+                if k.endswith(')'): # we don't care about mutations on sample leafs
+                    s = k.split()
+                    if len(s) == 1:
+                        myNode = int((s[0])[1:-1])
+                        mutSplit = []
+                    else:
+                        myNode = int((s[1])[1:-1])
+                        mutSplit = s[0].split(',')
+                    if not myNode in nodeToMuts:
+                        nodeToMuts[myNode] = []
+                        for m in pastMuts:
+                            nodeToMuts[myNode].append(m)
+                        for m in mutSplit:
+                            nodeToMuts[myNode].append(m)
+                    for m in mutSplit:
+                        pastMuts.append(m)
+
+
+    goodNodes = {}
+    with open(myL) as f:
+        for line in f:
+            splitLine = (line.strip()).split('\t')
+            if splitLine[0].isdigit() and int(splitLine[1]) >= myT:
+                goodNodes[int(splitLine[0])] = True
+
+    myOutMSA = ''
+    myOutString = ''
+    for n in nodeToMuts:
+        if n in goodNodes and n != 1: # don't use the root -- it has no mutations and is not useful for our purposes
+            myOutMSA += '>NODE'+str(n)+'\n'+makeChanges(myReference, nodeToMuts[n])+'\n'
+            myOutString += str(n)+'\t'+','.join(nodeToMuts[n])+'\n'
+    open('internal_nodes.msa.fa','w').write(myOutMSA)
+    open('nodeToMuts.txt','w').write(myOutString)
+
+
+##########################
+#### HELPER FUNCTIONS ####
+##########################
+
+def stripEach(myList):
+    myReturn = []
+    for k in myList:
+        myReturn.append(k.strip())
+    return(myReturn)
+
+def makeChanges(ref, muts):
+    myReturn = []
+    for k in list(ref):
+        myReturn.append(k)
+    for k in muts:
+        myCoord = int(k[1:-1])-1
+        if not myReturn[myCoord] == k[0]:
+            print(k, myReturn[myCoord])
+        myReturn[myCoord] = str(k[-1])
+        if not myReturn[myCoord] == k[-1]:
+            print(k, myReturn[myCoord])
+    return(''.join(myReturn))
+
+def getPos(myInds, intLineNumToPos):
+    myReturn = []
+    for k in myInds:
+        myReturn.append(intLineNumToPos[k])
+    return(myReturn)
+
+def joiner(entry):
+    newList = []
+    for k in entry:
+        newList.append(str(k))
+    return '\t'.join(newList)
+
+def joinerU(entry):
+    newList = []
+    for k in entry:
+        newList.append(str(k))
+    return '_'.join(newList)
+
+def joinerC(entry):
+    newList = []
+    for k in entry:
+        newList.append('"'+str(k)+'"')
+    return ','.join(newList)
+
+#########################
+##### FUNCTION CALL #####
+#########################
+
+def main(myCommandLine=None):
+    """
+    Initializes a CommandLine object and passes the provided 
+    arguments into a new fileConverter object and calls main method.
+    """
+    myCommandLine = CommandLine()
+
+    # Necessary files:
+    if myCommandLine.args.samples:
+        myS = myCommandLine.args.samples
+    if myCommandLine.args.leaves:
+        myL = myCommandLine.args.leaves
+    if myCommandLine.args.threshold:
+        myT = int(myCommandLine.args.threshold)
+    else:
+        myT = 10
+    if myCommandLine.args.ref:
+        myR = myCommandLine.args.ref
+
+    getMutationsFile(myS, myL, myT, myR)
+
+
+if __name__ == "__main__":
+    """
+    Calls main when program is run by user.
+    """
+    main();
+    raise SystemExit
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+