# HG changeset patch # User proteore # Date 1511739403 18000 # Node ID 6a45ccfc0e4cf1f2beed92d774df06fbaf74afab planemo upload commit abb24d36c776520e73220d11386252d848173697-dirty diff -r 000000000000 -r 6a45ccfc0e4c README.rst --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/README.rst Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,62 @@ +Wrapper for Filter out keywords and/or numerical values tool +============================================================ + +**Authors** + +T.P. Lien Nguyen, Florence Combes, Yves Vandenbrouck CEA, INSERM, CNRS, Grenoble-Alpes University, BIG Institute, FR +Sandra Dérozier, Olivier Rué, Christophe Caron, Valentin Loux INRA, Paris-Saclay University, MAIAGE Unit, Migale Bioinformatics platform + +This work has been partially funded through the French National Agency for Research (ANR) IFB project. + +Contact support@proteore.org for any questions or concerns about the Galaxy implementation of this tool. + +------------------------------------------------------------ + +This tool allows to remove unneeded data (e.g. contaminants, non-significant values) from a proteomics results file (e.g. MaxQuant or Proline output). + +**For each row, if there are more than one protein IDs/protein names/gene names, only the first one will be considered in the output** + +**Filter the file by keywords** + +Several options can be used. For each option, you can fill in the field or upload a file which contains the keywords. + +- If you choose to fill in the field, the keywords should be separated by ":", for example: A8K2U0:Q5TA79:O43175 + +- If you choose to upload a file in a text format in which each line is a keyword, for example: + + REV + + TRYP_PIG + + ALDOA_RABBIT + +**The line that contains these keywords will be eliminated from input file.** + +**Keywords search can be applied by performing either exact match or partial one by using the following option** + +- If you choose **Yes**, only the fields that contains exactly the same content will be removed. + +- If you choose **No**, all the fields containing the keyword will be removed. + +For example: + +**Yes** option (exact match) selected using the keyword "kinase": only lines which contain exactly "kinase" is removed. + +**No** option (partial match) for "kinase": not only lines which contain "kinase" but also lines with "alpha-kinase" (and so on) are removed. + +**Filter the file by values** + +You can choose to use one or more options (e.g. to filter out peptides of low intensity value, by q-value, etc.). + +* For each option, you can choose between "=", ">", ">=", "<" and "<=", then enter the value to filter and specify the column to apply that option. + +**Output** + +The tool will produce 2 output files. + +* A text file containing the resulting filtered input file. + +* A text file containing the rows removed from the input file. + + + diff -r 000000000000 -r 6a45ccfc0e4c filter_kw_val.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/filter_kw_val.py Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,180 @@ +import argparse +import re + + +def options(): + parser = argparse.ArgumentParser() + parser.add_argument("-i", "--input", help="Input file", required=True) + parser.add_argument("-m", "--match", help="Exact macth") + parser.add_argument("--kw", nargs="+", action="append", help="") # + parser.add_argument("--kw_file", nargs="+", action="append", help="") + parser.add_argument("--value", nargs="+", action="append", help="") + parser.add_argument("-o", "--output", default="output.txt") + parser.add_argument("--trash_file", default="trash_MQfilter.txt") + + args = parser.parse_args() + + filters(args) + + # python filter2.py -i "/projet/galaxydev/galaxy/tools/proteore_uc1/proteinGroups_Maud.txt" --protein_IDs "A2A288:A8K2U0" --peptides 2 "=" -o "test-data/output_MQfilter.txt" + + +def isnumber(format, n): + float_format = re.compile("^[\-]?[1-9][0-9]*\.?[0-9]+$") + int_format = re.compile("^[\-]?[1-9][0-9]*$") + test = "" + if format == "int": + test = re.match(int_format, n) + elif format == "float": + test = re.match(float_format, n) + if test: + return True + else: + return False + +def filters(args): + MQfilename = args.input.split(",")[0] + header = args.input.split(",")[1] + MQfile = readMQ(MQfilename) + results = [MQfile, None] + + if args.kw: + keywords = args.kw + for k in keywords: + results = filter_keyword(results[0], header, results[1], k[0], k[1], k[2]) + if args.kw_file: + key_files = args.kw_file + for kf in key_files: + ids = readOption(kf[0]) + results = filter_keyword(results[0], header, results[1], ids, kf[1], kf[2]) + if args.value: + for v in args.value: + if isnumber("float", v[0]): + results = filter_value(results[0], header, results[1], v[0], v[1], v[2]) + else: + raise ValueError("Please enter a number in filter by value") + + # Write results to output + output = open(args.output, "w") + output.write("".join(results[0])) + output.close() + + # Write deleted lines to trash_file + trash = open(args.trash_file, "w") + #print("".join(results[1])) + trash.write("".join(results[1])) + trash.close() + +def readOption(filename): + f = open(filename, "r") + file = f.read() + #print(file) + filter_list = file.split("\n") + #print(filter_list) + filters = "" + for i in filter_list: + filters += i + ":" + filters = filters[:-1] + #print(filters) + return filters + +def readMQ(MQfilename): + # Read MQ file + mqfile = open(MQfilename, "r") + mq = mqfile.readlines() + # Remove empty lines (contain only space or new line or "") + [mq.remove(blank) for blank in mq if blank.isspace() or blank == ""] + return mq + +def filter_keyword(MQfile, header, filtered_lines, ids, ncol, match): + mq = MQfile + if isnumber("int", ncol.replace("c", "")): + id_index = int(ncol.replace("c", "")) - 1 #columns.index("Majority protein IDs") + else: + raise ValueError("Please specify the column where you would like to apply the filter with valid format") + + ids = ids.upper().split(":") + [ids.remove(blank) for blank in ids if blank.isspace() or blank == ""] + + if header == "true": + header = mq[0] + content = mq[1:] + else: + header = "" + content = mq[:] + + if not filtered_lines: # In case there is already some filtered lines from other filters + filtered_lines = [] + if header != "": + filtered_lines.append(header) + + for line in content: + id_inline = line.split("\t")[id_index].replace('"', "").split(";") + one_id_line = line.replace(line.split("\t")[id_index], id_inline[0]) # Take only first IDs + + if match != "false": + # Filter protein IDs + if any (pid.upper() in ids for pid in id_inline): + #ids = prot_ids.split(":") + #print(prot_ids.split(":")) + #if prot_id in ids: + filtered_lines.append(one_id_line) + mq.remove(line) + else: + mq[mq.index(line)] = one_id_line + else: + if any (ft in pid.upper() for pid in id_inline for ft in ids): + filtered_lines.append(one_id_line) + mq.remove(line) + else: + mq[mq.index(line)] = one_id_line + return mq, filtered_lines + +def filter_value(MQfile, header, filtered_prots, filter_value, ncol, opt): + mq = MQfile + if ncol and isnumber("int", ncol.replace("c", "")): #"Gene names" in columns: + index = int(ncol.replace("c", "")) - 1 #columns.index("Gene names") + else: + raise ValueError("Please specify the column where you would like to apply the filter with valid format") + + if header == "true": + header = mq[0] + content = mq[1:] + else: + header = "" + content = mq[:] + + if not filtered_prots: # In case there is already some filtered lines from other filters + filtered_prots = [] + if header != "": + filtered_prots.append(header) + + for prot in content: + filter_value = float(filter_value) + pep = prot.split("\t")[index].replace('"', "") + if pep.replace(".", "", 1).isdigit(): + if opt == "<": + if not float(pep) < filter_value: + filtered_prots.append(prot) + mq.remove(prot) + elif opt == "<=": + if not float(pep) <= filter_value: + filtered_prots.append(prot) + mq.remove(prot) + elif opt == ">": + #print(prot.number_of_prots, filter_value, int(prot.number_of_prots) > filter_value) + if not float(pep) > filter_value: + filtered_prots.append(prot) + mq.remove(prot) + elif opt == ">=": + if not float(pep) >= filter_value: + filtered_prots.append(prot) + mq.remove(prot) + else: + if not float(pep) == filter_value: + filtered_prots.append(prot) + mq.remove(prot) + return mq, filtered_prots #output, trash_file + +if __name__ == "__main__": + options() diff -r 000000000000 -r 6a45ccfc0e4c filter_kw_val.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/filter_kw_val.xml Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,202 @@ + + Filter a file by keywords or values + + + + + + " + #else if $val.v.val == "Equal or higher" + $val.v.equal_higher "$val.v.ncol" ">=" + #else if $val.v.val == "Lower" + $val.v.lower "$val.v.ncol" "<" + #else + $val.v.equal_lower "$val.v.ncol" "<=" + #end if + #end if + #end for + + ]]> + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + ", ">=", "<" and "<=", then enter the value to filter and specify the column to apply that option. + +**Output** + +The tool will produce 2 output files. + +* A text file containing the resulting filtered input file. + +* A text file containing the rows removed from the input file. + +----- + +.. class:: infomark + +**Authors** + +T.P. Lien Nguyen, Florence Combes, Yves Vandenbrouck CEA, INSERM, CNRS, Grenoble-Alpes University, BIG Institute, FR +Sandra Dérozier, Olivier Rué, Christophe Caron, Valentin Loux INRA, Paris-Saclay University, MAIAGE Unit, Migale Bioinformatics platform + +This work has been partially funded through the French National Agency for Research (ANR) IFB project. + +Contact support@proteore.org for any questions or concerns about the Galaxy implementation of this tool. + + ]]> + + + diff -r 000000000000 -r 6a45ccfc0e4c test-data/UnipIDs.txt --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/UnipIDs.txt Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,25 @@ +P04637 +P08246 +P63244 +P10275 +P00533 +Q14524 +P05067 +P35555 +P35222 +O95273 +P00451 +P38398 +Q05086 +Q12802 +P68871 +P04585 +Q96EB6 +Q9NYL2 +P31749 +P01137 +Q5S007 +Q08379 +P02649 +P35498 +P12931 diff -r 000000000000 -r 6a45ccfc0e4c test-data/filter_keywords_values_output.txt --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/filter_keywords_values_output.txt Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,14 @@ +P08246 2 B0 +P63244 1.5 C1 +Q14524 3.5 D1 +P05067 1 B3 +P00451 2 B2 +P38398 5 B4 +Q12802 3 D5 +P68871 1.5 B4 +P04585 2.5 D3 +Q9NYL2 1 B1 +P01137 5 B6 +Q5S007 8 D4 +Q08379 2 C4 +P35498 1 C5 diff -r 000000000000 -r 6a45ccfc0e4c test-data/filter_keywords_values_removed.txt --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/test-data/filter_keywords_values_removed.txt Sun Nov 26 18:36:43 2017 -0500 @@ -0,0 +1,11 @@ +P04637 1 A0 +P10275 3 A2 +P00533 2 A3 +O95273 1.1 A4 +P31749 3 A1 +P12931 3 A5 +P35555 0 C0 +P35222 0.9 D2 +Q05086 0 C2 +Q96EB6 0 C3 +P02649 0 B5