Mercurial > repos > chmaramis > irprofiler
changeset 0:0e37e5b73273 draft
Initial commit
| author | chmaramis |
|---|---|
| date | Fri, 30 Mar 2018 07:22:29 -0400 |
| parents | |
| children | acaa8e8a0b88 |
| files | README.rst clonotype_computation.py clonotype_computation.xml data_filtering.py data_filtering.xml exclusive_clonotype_computation.py exclusive_clonotype_computation.xml gene_comparison.py gene_comparison.xml gene_computation.py gene_computation.xml public_clonotype_computation.py public_clonotype_computation.xml |
| diffstat | 13 files changed, 1096 insertions(+), 0 deletions(-) [+] |
line wrap: on
line diff
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/README.rst Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,9 @@ +About +----- + +IRProfiler is a Galaxy toolbox for immunogenetic repertoire profiling. It is made available as supplementary material for the article *IRProfiler - A Software Toolbox for High Throughput Immune Receptor Profiling*, authored by C. Maramis et al. and submitted for possible publication to `BMC Bioinformatics <https://bmcbioinformatics.biomedcentral.com>`_. + +Tools +----- + +IRProfiler consists of 6 tools. All tools require the Pandas library (v 0.19).
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/clonotype_computation.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,90 @@ +# -*- coding: utf-8 -*- +""" +Created on Sat Mar 24 14:00:24 2018 + +@author: chmaramis +""" + +from __future__ import division +import numpy as np +from pandas import * +import functools as ft +import sys +import time + +frm = lambda x,y: '{r}/{l}'.format(r=x,l=y) + +clono_def = {'CDR3': ['AA JUNCTION'], + 'VCDR3': ['V-GENE','AA JUNCTION'], + 'JCDR3': ['J-GENE','AA JUNCTION'], + 'VJCDR3': ['V-GENE','J-GENE','AA JUNCTION'], + 'VDJCDR3': ['V-GENE','D-GENE','J-GENE','AA JUNCTION']} + + +def clonotypeComputation(inp_name, clono, out1, t10n, fname): + + clono_comps = clono_def[clono] + + frame = DataFrame() + tp = read_csv(inp_name, iterator=True, chunksize=5000,sep='\t', index_col=0 ) + frame = concat([chunk for chunk in tp]) + + + grouped = frame.groupby(clono_comps) + x=grouped.size() + x1=DataFrame(list(x.index), columns=clono_comps) + x1['Reads']=x.values + total = sum(x1['Reads']) + #x1['Reads/Total'] = ['{r}/{l}'.format(r=pr , l = total) for pr in x1['Reads']] + x1['Reads/Total'] = x1['Reads'].map(ft.partial(frm, y=total)) + x1['Frequency %'] = (100*x1['Reads']/total).map('{:.4f}'.format) + + final = x1.sort_values(by = ['Reads'] , ascending = False) + + final.index=range(1,len(final)+1) + final.to_csv(out1 , sep = '\t') + + numofclono = len(final) + clust = len(final[final['Reads'] > 1]) + sing = len (final[final['Reads'] == 1]) + top10 = final[clono_comps + ['Frequency %']].head(10) + top10.to_csv(t10n , sep = '\t') + + summary = [[clono]] + summary.append([', '.join([top10[c].values[0] for c in clono_comps])]) + summary.append([top10['Frequency %'].values[0]]) + summary.append([numofclono]) + summary.append([clust,'{:.4f}'.format(100*clust/numofclono)]) + summary.append([sing,'{:.4f}'.format(100*sing/numofclono)]) + + ind = ['Clonotype Definition', 'Dominant Clonotype', 'Frequency', 'Number of Clonotypes' , 'Expanding Clonotypes','Singletons'] + spl = fname.split('_') + col = [spl[0],'%'] + + frsum = DataFrame(summary,index = ind, columns = col) + + return frsum + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + inp_name = sys.argv[1] + clono = sys.argv[2] + out1 = sys.argv[3] + t10n = sys.argv[4] + sname = sys.argv[5] + fname = sys.argv[6] + + # Execute basic function + frsum = clonotypeComputation(inp_name, clono, out1, t10n, fname) + + # Save output to CSV files + if not frsum.empty: + frsum.to_csv(sname, sep = '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start)) \ No newline at end of file
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/clonotype_computation.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,25 @@ +<tool id="clonoComput" name="Clonotype Diversity & Expression" version="0.9"> + <description>Compute clonotype diversity and expression from filtered file</description> + <requirements> + <requirement type="package" version="0.19">pandas</requirement> + </requirements> + <command interpreter="python">clonotype_computation.py $input $clonotype $clonos_file $top10_file $summary_file ${input.name}</command> + <inputs> + <param name="clonotype" type="select" label="Clonotype definition"> + <option value="CDR3">CDR3</option> + <option value="VCDR3">V+CDR3</option> + <option value="JCDR3">J+CDR3</option> + <option value="VJCDR3">V+J+CDR3</option> + <option value="VDJCDR3">V+D+J+CDR3</option> + </param> + <param format="tabular" name="input" type="data" label="Filtered-in File" /> + </inputs> + <outputs> + <data name="clonos_file" format="tabular" label="${input.name}_clonotypesAll" /> + <data name="top10_file" format="tabular" label="${input.name}_clonotypesTop10" /> + <data name="summary_file" format="tabular" label="${input.name}_clonotypesSummary" /> + </outputs> + <help> +Coming soon + </help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_filtering.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,449 @@ +# -*- coding: utf-8 -*- +""" +Created on Wed Sep 4 18:41:42 2013 + +@author: chmaramis +""" + +from __future__ import division +import string as strpy +import numpy as np +from pandas import * +from numpy import nan as NA +import time +import sys + + +def filter_condition_AAjunction(x): + x= x.strip() + if ' ' in x: + return x.split(' ')[0] + else: + return x + +#-----------frame creation--------------------- +def dataFiltering(inp,cells,psorf,con,prod,CF,Vper,Vgene,laa1,laa2,conaa,Jgene,Dgene,fname): + + try: + path=inp + frame = DataFrame() + seqlen = [] + head = [] + tp = read_csv(path, iterator=True, chunksize=5000,sep='\t', index_col=0 ) + frame = concat([chunk for chunk in tp]) + + frcol = list(frame.columns) + #print frcol[-1] + if 'Unnamed' in frcol[-1]: + del frcol[-1] + frame=frame[frcol] + + frame.index = range(1,len(frame)+1) + + head.append('Total reads of raw data') + seqlen.append(len(frame)) + + #------------drop nulls-------------------- + filtered = DataFrame() + filtall = DataFrame() + summ_df = DataFrame() + filtered = frame[isnull(frame['AA JUNCTION']) | isnull(frame['V-GENE and allele'])] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = "NoResults" + frame = frame[frame['AA JUNCTION'].notnull()] + frame = frame[frame['V-GENE and allele'].notnull()] + + head.append('Not Null CDR3/V') + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + filtered = DataFrame() + + if psorf.startswith('y') or psorf.startswith('Y'): + + cc0=np.array(frame['V-GENE and allele'].unique()) + + + for x in cc0: + x1=x.split('*') + try: + if (x1[1].find('P')>-1) or (x1[1].find('ORF')>-1): + filtered = filtered.append(frame[frame['V-GENE and allele'] == x]) + frame['V-GENE and allele']=frame['V-GENE and allele'].replace(x,NA) + elif x.find('or')>-1: + posa=x.count('or') + x2=x.split('or') + x4='' + genelist=[] + for cnt in range(0, posa+1): + x3=x2[cnt].split('*') + x3[0]=x3[0].strip()#kobei ta space + k=x3[0].split(' ')# holds only TRBV + if cnt==0: + genelist.append(k[1]) + x4+=k[1] + elif ((str(k[1]) in genelist) == False) & (x3[1].find('P')==-1):# check for P in x3 + genelist.append(k[1]) + x4+=' or ' + x4+=k[1] + x3=None + k1=None + genelist=None + + frame['V-GENE and allele']=frame['V-GENE and allele'].replace(x,x4) + + else: + s=x1[0].split(' ') + frame['V-GENE and allele']=frame['V-GENE and allele'].replace(x,s[1]) + except IndexError as e: + print('V-gene is already been formed') + continue + + x=None + x1=None + s=None + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'P or ORF' + frame = frame[frame['V-GENE and allele'].notnull()] + + head.append('Functional TRBV') + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + filtered = DataFrame() + + + + #------------FILTERING for data quality-------------------- + if con.startswith('y') or con.startswith('Y'): + filtered = frame [frame['AA JUNCTION'].str.contains('X') | + frame['AA JUNCTION'].str.contains('#') | + frame['AA JUNCTION'].str.contains('[*]')] + + + + frame = frame [~frame['AA JUNCTION'].str.contains('X') & + ~frame['AA JUNCTION'].str.contains('#') & + ~frame['AA JUNCTION'].str.contains('[*]') ] + + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'X,#,*' + head.append('Not Containing X,#,*') + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + filtered = DataFrame() + + + + if prod.startswith('y') or prod.startswith('Y'): + filtered = frame[~frame['Functionality'].str.startswith('productive')] + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'not productive' + + + frame=frame[frame['Functionality'].str.startswith('productive')] + + head.append('Productive') + head.append('filter out') + seqlen.append(len(frame)) + + seqlen.append(len(filtered)) + + + frame['AA JUNCTION'] = frame['AA JUNCTION'].map(filter_condition_AAjunction) + + if CF.startswith('y') or CF.startswith('Y'): + if cells == 'TCR': + filtered = DataFrame() + filtered = frame[~frame['AA JUNCTION'].str.startswith('C') | + ~frame['AA JUNCTION'].str.endswith('F')] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'Not C..F' + + frame = frame[frame['AA JUNCTION'].str.startswith('C') & + frame['AA JUNCTION'].str.endswith('F')] + + head.append('CDR3 landmarks C-F') + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + filtered = DataFrame() + elif cells == 'BCR': + filtered = DataFrame() + filtered = frame[~frame['AA JUNCTION'].str.startswith('C') | + ~frame['AA JUNCTION'].str.endswith('W')] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'Not C..W' + + frame = frame[frame['AA JUNCTION'].str.startswith('C') & + frame['AA JUNCTION'].str.endswith('W')] + + head.append('CDR3 landmarks C-W') + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + filtered = DataFrame() + else: + print('TCR or BCR type') + + + filtered = DataFrame() + + filtered = frame[frame['V-REGION identity %'] < Vper] + + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'identity < {iden}%'.format(iden = Vper) + + frame=frame[frame['V-REGION identity %']>= Vper] + head.append('Identity >= {iden}%'.format(iden = Vper)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + head.append('Total filter out A') + head.append('Total filter in A') + seqlen.append(len(filtall)) + seqlen.append(len(frame)) + + ############################### + if Vgene != 'null': + + filtered = DataFrame() + + filtered = frame[frame['V-GENE and allele'] != Vgene] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'V-GENE != {} '.format(Vgene) + + + frame = frame[frame['V-GENE and allele'] == Vgene] + + + + head.append('V-GENE = {} '.format(Vgene)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + + + ############################### + if (laa1 != 'null') or (laa2 != 'null'): + if int(laa2) == 0: + low = int(laa1) + high = 100 + elif int(laa1) > int(laa2): + low = int(laa2) + high = int(laa1) + else: + low = int(laa1) + high = int(laa2) + + filtered = DataFrame() + criteria = frame['AA JUNCTION'].apply(lambda row: (len(row)-2) < low) + criteria2 = frame['AA JUNCTION'].apply(lambda row: (len(row)-2) > high) + filtered = frame[criteria | criteria2] + + filtall = filtall.append(filtered) + if int(laa2)==0: + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'CDR3 length not bigger than {}'.format(low) + else: + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'CDR3 length not from {} to {}'.format(low,high) + + criteria3 = frame['AA JUNCTION'].apply(lambda row: (len(row)-2) >= low) + criteria4 = frame['AA JUNCTION'].apply(lambda row: (len(row)-2) <= high) + frame = frame[criteria3 & criteria4] + + if int(laa2)==0: + head.append('CDR3 length bigger than {}'.format(low)) + else: + head.append('CDR3 length from {} to {} '.format(low,high)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + ############################### + if conaa != 'null': + if conaa.islower(): + conaa = conaa.upper() + filtered = DataFrame() + + filtered = frame[~frame['AA JUNCTION'].str.contains(conaa)] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'CDR3 not containing {}'.format(conaa) + + frame = frame[frame['AA JUNCTION'].str.contains(conaa) ] + + head.append('CDR3 containing {}'.format(conaa)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + + + + #####------------keep the small J gene name-------------------- + #frame['J-GENE and allele'] = frame['J-GENE and allele'].map(filter_condition_Jgene) + cc2=np.array(frame['J-GENE and allele'].unique()) + + for x in cc2: + try: + if notnull(x): + x1=x.split('*') + # print(x) + # print (x1[0]) + trbj=x1[0].split(' ') + frame['J-GENE and allele']=frame['J-GENE and allele'].replace(x,trbj[1]) + except IndexError as e: + print('J-Gene has been formed') + + + + x=None + x1=None + + + #------------keep the small D gene name-------------------- + cc1=np.array(frame['D-GENE and allele'].unique()) + for x in cc1: + try: + if notnull(x): + x1=x.split('*') + trbd=x1[0].split(' ') + frame['D-GENE and allele']=frame['D-GENE and allele'].replace(x,trbd[1]) + else: + frame['D-GENE and allele']=frame['D-GENE and allele'].replace(x,'none') + except IndexError as e: + print('D-gene has been formed') + + + x=None + x1=None + + + if Jgene != 'null': + + filtered = DataFrame() + + filtered = frame[frame['J-GENE and allele'] != Jgene] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'J-GENE not {} '.format(Jgene) + + + frame = frame[frame['J-GENE and allele'] == Jgene] + + + + head.append('J-GENE = {} '.format(Jgene)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + + + if Dgene != 'null': + + filtered = DataFrame() + + filtered = frame[frame['D-GENE and allele'] != Dgene] + + filtall = filtall.append(filtered) + if len(filtall) > 0: + filtall.loc[filtered.index,'Reason'] = 'D-GENE not {} '.format(Dgene) + + + frame = frame[frame['D-GENE and allele'] == Dgene] + + + + head.append('D-GENE = {} '.format(Dgene)) + head.append('filter out') + seqlen.append(len(frame)) + seqlen.append(len(filtered)) + + + head.append('Total filter out') + head.append('Total filter in') + seqlen.append(len(filtall)) + seqlen.append(len(frame)) + summ_df = DataFrame(index = head) + col = fname + + summ_df[col] = seqlen + frame=frame.rename(columns = {'V-GENE and allele':'V-GENE', + 'J-GENE and allele':'J-GENE','D-GENE and allele':'D-GENE'}) + + + frcol.append('Reason') + + filtall = filtall[frcol] + + #--------------out CSV--------------------------- + frame.index = range(1,len(frame)+1) + if not summ_df.empty: + summ_df['%'] = (100*summ_df[summ_df.columns[0]]/summ_df[summ_df.columns[0]][summ_df.index[0]]).map(('{:.4f}'.format)) + return(frame,filtall,summ_df) + except KeyError as e: + print('This file has no ' + str(e) + ' column') + return(frame,filtall,summ_df) + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + inp = sys.argv[1] + cells = sys.argv[2] + psorf = sys.argv[3] + con = sys.argv[4] + prod = sys.argv[5] + CF = sys.argv[6] + Vper = float(sys.argv[7]) + Vgene = sys.argv[8] + laa1 = sys.argv[9] + conaa = sys.argv[10] + filterin = sys.argv[11] + filterout = sys.argv[12] + Sum_table = sys.argv[13] + Jgene = sys.argv[14] + Dgene = sys.argv[15] + laa2 = sys.argv[16] + fname = sys.argv[17] + + # Execute basic function + fin,fout,summ = dataFiltering(inp,cells,psorf,con,prod,CF,Vper,Vgene,laa1,laa2,conaa,Jgene,Dgene,fname) + + # Save output to CSV files + if not summ.empty: + summ.to_csv(Sum_table, sep = '\t') + if not fin.empty: + fin.to_csv(filterin , sep = '\t') + if not fout.empty: + fout.to_csv(filterout, sep= '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start)) +
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/data_filtering.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,103 @@ +<tool id="dataFilter" name="Data Filtering" version="0.9"> + <description>Filter IMGT Summary dataset</description> + <requirements> + <requirement type="package" version="0.19">pandas</requirement> + </requirements> + <command interpreter="python">data_filtering.py $input $TCR_or_BCR $Vfun $spChar $prod $delCF $threshold $Vg.Vgid $clen.cdr3len1 $cdp.cdr3part $filterin_file $filterout_file $summary_file $Jg.Jgid $Dg.Dgid $clen.cdr3len2 $process_id + </command> + <inputs> + <param format="txt" name="input" type="data" label="IMGT Summary Output" /> + <param format="txt" name="process_id" type="text" label="Process ID" /> + <param name="TCR_or_BCR" type="select" label="T-cell or B-cell option"> + <option value="TCR">T-cell</option> + <option value="BCR">B-cell</option> + </param> + <param name="Vfun" type="select" label="Only Take Into Account Fuctional V-GENE? "> + <option value="y">yes</option> + <option value="n">no</option> + </param> + <param name="spChar" type="select" label="Only Take Into Account CDR3 with no Special Characters (X,*,#)? "> + <option value="y">yes</option> + <option value="n">no</option> + </param> + <param name="prod" type="select" label="Only Take Into Account Productive Sequences? "> + <option value="y">yes</option> + <option value="n">no</option> + </param> + <param name="delCF" type="select" label="Only Take Into Account CDR3 with valid start/end landmarks? "> + <option value="y">yes</option> + <option value="n">no</option> + </param> + <param name="threshold" type="float" size="3" value="0" min="0" max="100" label="V-REGION identity %" /> + <conditional name="Vg"> + <param name="Vg_select" type="select" label="Select Specific V gene?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param format="txt" name="Vgid" type="text" label="Type V gene" /> + </when> + <when value="n"> + <param name="Vgid" type="hidden" value="null" /> + </when> + </conditional> + <conditional name="Jg"> + <param name="Jg_select" type="select" label="Select Specific J gene?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param format="txt" name="Jgid" type="text" label="Type J gene" /> + </when> + <when value="n"> + <param name="Jgid" type="hidden" value="null" /> + </when> + </conditional> + <conditional name="Dg"> + <param name="Dg_select" type="select" label="Select Specific D gene?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param format="txt" name="Dgid" type="text" label="Type D gene" /> + </when> + <when value="n"> + <param name="Dgid" type="hidden" value="null" /> + </when> + </conditional> + <conditional name="clen"> + <param name="clen_select" type="select" label="Select CDR3 length range?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param name="cdr3len1" type="integer" size="3" value="0" min="0" max="100" label="CDR3 Length Lower Threshold" /> + <param name="cdr3len2" type="integer" size="3" value="0" min="0" max="100" label="CDR3 Length Upper Threshold" /> + </when> + <when value="n"> + <param name="cdr3len1" type="hidden" value="null" /> + <param name="cdr3len2" type="hidden" value="null" /> + </when> + </conditional> + <conditional name="cdp"> + <param name="cdp_select" type="select" label="Only select CDR3 containing specific amino-acid sequence?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param format="txt" name="cdr3part" type="text" label="Type specific amino-acid sequence" /> + </when> + <when value="n"> + <param name="cdr3part" type="hidden" value="null" /> + </when> + </conditional> + </inputs> + <outputs> + <data name="filterin_file" format="tabular" label="${process_id}_filterin" /> + <data name="filterout_file" format="tabular" label="${process_id}_filterout" /> + <data name="summary_file" format="tabular" label="${process_id}_filterSummary" /> + </outputs> + <help> +This tool filters an IMGT Summary dataset based on a combination of criteria. + </help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/exclusive_clonotype_computation.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,70 @@ +# -*- coding: utf-8 -*- +""" +Created on Sat Mar 24 17:31:38 2018 + +@author: chmaramis +""" + +from __future__ import division +import numpy as np +from pandas import * +from numpy import nan as NA +import sys +import time + +clono_def = {'CDR3': ['AA JUNCTION'], + 'VCDR3': ['V-GENE','AA JUNCTION'], + 'JCDR3': ['J-GENE','AA JUNCTION']} + + +def exclusiveClonotypeComputation(inputs, clono, thres): + + clono_comps = clono_def[clono] + + vClono=DataFrame() + + # File A + cl = DataFrame() + cl = read_csv(inputs[0] , sep = '\t' , index_col = 0) + if (thres != 'null'): + cl = cl[cl['Reads'] > int(thres)] + vClono = cl + + # File B + cl = DataFrame() + cl = read_csv(inputs[2] , sep = '\t' , index_col = 0) + if (thres != 'null'): + cl = cl[cl['Reads'] > int(thres)] + cl.rename(columns={'Reads':'ReadsB'}, inplace=True) + vClono = vClono.merge(cl[clono_comps+['ReadsB']], how='left', on=clono_comps) + + vClono['ReadsB'].fillna(0, inplace=True) + + vClono = vClono[vClono['ReadsB'] == 0] + del vClono['ReadsB'] + + vClono.index = range(1,len(vClono)+1) + + return vClono + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + arg = sys.argv[4:] + clono = sys.argv[1] + output = sys.argv[2] + threshold = sys.argv[3] + + # Execute basic function + excl = exclusiveClonotypeComputation(arg, clono, threshold) + + # Save output to CSV files + if not excl.empty: + excl.to_csv(output , sep = '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start))
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/exclusive_clonotype_computation.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,40 @@ +<tool id="exclClonoComput" name="Exclusive Clonotypes" version="0.9"> +<description>Compute exclusive clonotypes between 2 repertoires</description> +<requirements> + <requirement type="package" version="0.19">pandas</requirement> +</requirements> +<command interpreter="python">exclusive_clonotype_computation.py "$clono" "$output_file" "$Th.thres" "$inputA" "$inputA.name" "$inputB" "$inputB.name" +</command> +<inputs> + <param name="clonotype" type="select" label="Clonotype definition"> + <option value="CDR3">CDR3</option> + <option value="VCDR3">V+CDR3</option> + <option value="JCDR3">J+CDR3</option> + </param> + <conditional name="Th"> + + <param name="thres_select" type="select" label="Remove CDR3 With Reads Fewer Than Threshold?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + + <when value="y"> + <param name="thres" type="integer" size="4" value="1" min="1" label="Keep CDR3 with Number of Reads more than"/> + </when> + + <when value="n"> + <param name="thres" type="hidden" value="null" /> + </when> + + </conditional> + <param format="txt" name="inputA" type="data" label="First Clonotypes File (A)"/> + <param format="txt" name="inputB" type="data" label="Second Clonotypes File (B)"/> +</inputs> + +<outputs> +<data format="tabular" name="output_file" label="exclusiveClonotypes"/> +</outputs> +<help> +Coming soon +</help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/gene_comparison.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,65 @@ +# -*- coding: utf-8 -*- +""" +Created on Sat Mar 24 17:45:09 2018 + +@author: chmaramis +""" + +from __future__ import division +import numpy as np +from pandas import * +from numpy import nan as NA +import sys +import time + +sw_clonos = lambda x: x.startswith('Clonotypes') +sw_freq = lambda x: x.startswith('Freq') +sw_gene = lambda x: x.endswith('GENE') + +def geneComparison(inputs): + + mer=DataFrame() + + for x in range(0,len(inputs),2): + + ini = read_csv(inputs[x] , sep = '\t' , index_col = 0) + + ini.drop(ini.columns[np.where(ini.columns.map(sw_clonos))[0]], axis=1, inplace=True) + + x1 = inputs[x+1].split('_') + ini.rename(columns={ini.columns[np.where(ini.columns.map(sw_freq))[0][0]]: x1[0]}, inplace=True) + + if mer.empty: + mer = DataFrame(ini) + else: + mer = merge(mer,ini, on=ini.columns[np.where(ini.columns.map(sw_gene))[0][0]] , how='outer') + + mer=mer.fillna(0) + mer['mean'] = mer.sum(axis=1)/(len(mer.columns)-1) + fr = 'mean' + + mer=mer.sort_values(by = fr,ascending=False) + mer[fr] = mer[fr].map('{:.4f}'.format) + mer.index = range(1,len(mer)+1) + + return mer + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + inputs = sys.argv[2:] + output = sys.argv[1] + + # Execute basic function + mer = geneComparison(inputs) + + # Save output to CSV files + if not mer.empty: + mer.to_csv(output , sep = '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start))
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/gene_comparison.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,23 @@ +<tool id="geneCompar" name="Gene Usage Comparison" version="0.9"> +<description>Compare gene usages from multiple repertoires</description> +<requirements> + <requirement type="package" version="0.19">pandas</requirement> +</requirements> +<command interpreter="python">gene_comparison.py "$output_file" +#for x in $rep_files + "$x.rpfile" + "$x.rpfile.name" +#end for +</command> +<inputs> +<repeat name="rep_files" title="Patient" min="2"> +<param name="rpfile" type="data" label="File of gene usage repertoire" format="tabular"/> +</repeat> +</inputs> +<outputs> +<data format="tabular" name="output_file" label="geneUsageComparison"/> +</outputs> +<help> +Coming soon +</help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/gene_computation.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,76 @@ +# -*- coding: utf-8 -*- +""" +Created on Fri Jun 20 14:58:08 2014 + +@author: chmaramis +""" + +from __future__ import division +import numpy as np +from pandas import * +import functools as ft +import sys +import time + +frm = lambda x,y: '{r}/{l}'.format(r=x,l=y) + +gene_options = {'V': 'V-GENE', + 'J': 'J-GENE'} + + +def geneComputation(inp_name, gene, fname): + + gene_full = gene_options[gene] + + df = DataFrame() + df = read_csv(inp_name, sep='\t', index_col=0 ) + #tp = read_csv(inp_name, iterator=True, chunksize=5000,sep='\t', index_col=0 ) + #df = concat([chunk for chunk in tp]) + + + vgroup = df.groupby([gene_full]) + vdi = vgroup.size() + rep = DataFrame(list(vdi.index), columns=[gene_full]) + rep['Clonotypes'] = vdi.values + #rep['Clonotypes/Total'] = ['{r}/{l}'.format(r=p , l = len(df)) for p in vdi.values] + rep['Clonotypes/Total'] = rep['Clonotypes'].map(ft.partial(frm, y=len(df))) + rep['Frequency %'] = (100*rep['Clonotypes']/len(df)).map('{:.4f}'.format) + + rep = rep.sort_values(by = ['Clonotypes'] , ascending = False) + rep.index = range(1,len(rep)+1) + + su = rep[[gene_full, 'Frequency %']].head(10) + spl = fname.split('_') + summdf = DataFrame([gene_full,su[gene_full].values[0],su['Frequency %'].values[0]], + index = ['Gene Family','Dominant Gene','Frequency'], columns = [spl[0]]) + summdf['%'] = '' + + return (rep, su, summdf) + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + inp_name = sys.argv[1] + gene = sys.argv[2] + outrep = sys.argv[3] + summ_rep = sys.argv[4] + summ_rep2 = sys.argv[5] + fname = sys.argv[6] + + # Execute basic function + rep, su, summdf = geneComputation(inp_name, gene, fname) + + # Save output to CSV files + if not rep.empty: + rep.to_csv(outrep, sep = '\t') + if not su.empty: + su.to_csv(summ_rep, sep = '\t') + if not summdf.empty: + summdf.to_csv(summ_rep2, sep = '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start))
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/gene_computation.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,22 @@ +<tool id="geneComput" name="Gene Usage" version="0.9"> + <description>Compute gene usage from clonotype file</description> + <requirements> + <requirement type="package" version="0.19">pandas</requirement> + </requirements> + <command interpreter="python">gene_computation.py $input $gene $usage_file $top10_file $summary_file ${input.name}</command> + <inputs> + <param name="gene" type="select" label="Gene family"> + <option value="V">V-Gene</option> + <option value="J">J-Gene</option> + </param> + <param format="tabular" name="input" type="data" label="Clonotype file" /> + </inputs> + <outputs> + <data name="usage_file" format="tabular" label="${input.name}_geneUsageAll" /> + <data name="top10_file" format="tabular" label="${input.name}_geneUsageTop10" /> + <data name="summary_file" format="tabular" label="${input.name}_geneUsageSummary" /> + </outputs> + <help> +Coming soon + </help> +</tool>
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/public_clonotype_computation.py Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,84 @@ +# -*- coding: utf-8 -*- +""" +Created on Sat Mar 24 17:18:09 2018 + +@author: chmaramis +""" + +from __future__ import division +import numpy as np +from pandas import * +from numpy import nan as NA +import sys +import time + +clono_def = {'CDR3': ['AA JUNCTION'], + 'VCDR3': ['V-GENE','AA JUNCTION'], + 'JCDR3': ['J-GENE','AA JUNCTION']} + + + +def publicClonotypeComputation(inputs, clono, thres): + + clono_comps = clono_def[clono] + + clono=DataFrame() + + for x in range(0,len(inputs),2): + cl = DataFrame() + cl = read_csv(inputs[x] , sep = '\t' , index_col = 0) + #tp = read_csv(inp_name, iterator=True, chunksize=5000,sep='\t', index_col=0 ) + #cl = concat([chunk for chunk in tp]) + + if (thres != 'null'): + cl = cl[cl['Reads'] > int(thres)] + + x1 = inputs[x+1].split('_') + + del cl['Reads'] + cl.columns = [cl.columns[0], cl.columns[1], x1[0]+' '+cl.columns[2], x1[0]+' Relative '+cl.columns[3]] + + if clono.empty: + clono = cl + else: + clono = clono.merge(cl, how='outer', on=clono_comps) + + + col = clono.columns + freqs = col.map(lambda x: 'Frequency' in x) + reads = col.map(lambda x: 'Reads/Total' in x) + + clono[col[freqs]] = clono[col[freqs]].fillna(0) + clono[col[reads]] = clono[col[reads]].fillna('0/*') + + clono['Num of Patients']= clono[col[freqs]].apply(lambda x: np.sum(x != 0), axis=1) + + clono = clono[clono['Num of Patients'] > 1] + + clono.index = range(1,len(clono)+1) + + return clono + + +if __name__ == '__main__': + + start=time.time() + + # Parse input arguments + arg = sys.argv[4:] + clono = sys.argv[1] + output = sys.argv[2] + thres = sys.argv[3] + + + + # Execute basic function + mer = publicClonotypeComputation(arg, clono, thres) + + # Save output to CSV files + if not mer.empty: + mer.to_csv(output , sep = '\t') + + # Print execution time + stop=time.time() + print('Runtime:' + str(stop-start))
--- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/public_clonotype_computation.xml Fri Mar 30 07:22:29 2018 -0400 @@ -0,0 +1,40 @@ +<tool id="pubClonoComput" name="Public Clonotypes" version="0.9"> + <description>Compute public clonotypes from multiple repertoires</description> + <requirements> + <requirement type="package" version="0.19">pandas</requirement> + </requirements> + <command interpreter="python">public_clonotype_computation.py "$clonotype" "$output_file" "$Th.thres" +#for x in $clono_files + "$x.clfile" + "$x.clfile.name" +#end for +</command> + <inputs> + <param name="clonotype" type="select" label="Clonotype definition"> + <option value="CDR3">CDR3</option> + <option value="VCDR3">V+CDR3</option> + <option value="JCDR3">J+CDR3</option> + </param> + <conditional name="Th"> + <param name="thres_select" type="select" label="Remove Clonotypes With Reads Fewer Than Threshold?"> + <option value="y">Yes</option> + <option value="n" selected="true">No</option> + </param> + <when value="y"> + <param name="thres" type="integer" size="4" value="1" min="1" label="Keep Clonotypes with Number of Reads more than" /> + </when> + <when value="n"> + <param name="thres" type="hidden" value="null" /> + </when> + </conditional> + <repeat name="clono_files" title="Files to be append" min="2"> + <param name="clfile" type="data" label="Clonotype_File" format="tabular" /> + </repeat> + </inputs> + <outputs> + <data format="tabular" name="output_file" label="publicClonotypes" /> + </outputs> + <help> +Coming soon +</help> +</tool>