| Previous changeset 26:326aadaa3dd9 (2018-07-18) Next changeset 28:4ca9ce54642f (2018-07-18) |
|
Commit message:
Uploaded |
|
added:
cravat_convert/__pycache__/base_converter.cpython-36.pyc cravat_convert/__pycache__/vcf_converter.cpython-36.pyc cravat_convert/base_converter.py cravat_convert/cravat_convert.py cravat_convert/cravat_convert.xml cravat_convert/vcf_converter.py |
|
removed:
cravat_annotate/cravat_annotate.py cravat_annotate/cravat_annotate.xml |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_annotate/cravat_annotate.py --- a/cravat_annotate/cravat_annotate.py Wed Jul 18 10:18:43 2018 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 |
| [ |
| @@ -1,132 +0,0 @@ -""" -A galaxy wrapper for the /rest/service/query API endpoint on Cravat. -""" - -import requests -import json -import sys -import re -import argparse -from __future__ import print_function - - -# The endpoint that CravatQuerys are submitted to -endpoint = 'http://www.cravat.us/CRAVAT/rest/service/query' - - -# newline and delimiter values used in the output file -delimiter = "\t" -newline = "\n" - - -# Defualt indices for intepretting a cravat file's row of data in to a CravatQuery -cr_mapping = { - 'chromosome': 1, - 'position': 2, - 'strand': 3, - 'reference': 4, - 'alternate': 5 -} - - -# The neccessary attributes neeeded to submit a query. -query_keys = [ - 'chromosome', 'position', 'strand', 'reference', 'alternate' -] - - -# Expected response keys from server. Ordered in list so that galaxy output has uniform column ordering run-to-run. -# If cravat server returns additional keys, they are appended to and included in output. -ordered_keys = [ - "Chromosome", "Position", "Strand", "Reference base(s)", "Alternate base(s)", - "HUGO symbol", "S.O. transcript", "Sequence ontology protein change", "Sequence ontology", - "S.O. all transcripts", "gnomAD AF", "gnomAD AF (African)", "gnomAD AF (Amrican)", - "gnomAD AF (Ashkenazi Jewish)", "gnomAD AF (East Asian)", "gnomAD AF (Finnish)", - "gnomAD AF (Non-Finnish European)", "gnomAD AF (Other)", "gnomAD AF (South Asian)", - "1000 Genomes AF", "ESP6500 AF (average)", "ESP6500 AF (European American)", - "ESP6500 AF (African American)", "COSMIC transcript", "COSMIC protein change", - "COSMIC variant count [exact nucleotide change]", "cosmic_site_nt", "CGL driver class", - "TARGET", "dbSNP", "cgc_role", "cgc_inheritance", "cgc_tumor_type_somatic", - "cgc_tumor_type_germline", "ClinVar", "ClinVar disease identifier", "ClinVar XRef", - "GWAS Phenotype (GRASP)", "GWAS PMID (GRASP)", "Protein 3D variant" -] - - -def get_args(): - parser = argparse.ArgumentParser() - parser.add_argument('--input', - '-i', - required = True, - help='Input path to a cravat file for querying',) - parser.add_argument('--output', - '-o', - default = None, - help = 'Output path to write results from query') - return parser.parse_args() - - -def format_chromosome(chrom): - """ : Ensure chromosome entry is propely formatted for use as querying attribute. """ - if chrom[0:3] == 'chr': - return chrom - return 'chr' + str(chrom) - - -def get_query_string(row): - """ : From a row dict, return a query string for the Cravat server. - : The row dict is cravat headeres associated to their values of that row. - """ - return '_'.join([ row['chromosome'], row['position'], row['strand'], row['reference'], row['alternate'] ]) - - -def query(in_path, out_path): - """ : From a Cravat the file at in_path, query each line on the Cravat server. - : Write the response values to file at out_path. - """ - with open(in_path, 'r') as in_file, \ - open(out_path, 'w') as out_file: - for line in in_file: - try: - line = line.strip().split('\t') - # row is dict of cravat col headers assioted values in this line - row = { header: line[index] for header, index in cr_mapping.items() } - row['chromosome'] = format_chromosome(row['chromosome']) - query_string = get_query_string(row) - call = requests.get(endpoint, params={ 'mutation': query_string }) - if call.status_code != 200 or call.text == "": - raise requests.RequestException('Bad server response for query="{}". Respone code: "{}", Response Text: "{}"' - .format(query_string, call.status_code, call.text)) - json_response = json.loads(call.text) - # See if server returned additional json key-vals not expected in ordered_keys - for key in json_response: - if key not in ordered_keys: - ordered_keys.append(key) - # Write key in order of ordered_keys to standardize order of output columns - wrote = False - for key in ordered_keys: - if key in json_response: - val = json_response[key] - else: - val = None - # Standardize format for numeric values - try: - val = float(val) - val = format(val, ".4f") - except: - pass - if wrote: - out_file.write(delimiter) - out_file.write(str(val)) - wrote = True - out_file.write(newline) - except Exception as e: - print(e, file=sys.stderr) - continue - - -if __name__ == "__main__": - cli_args = get_args() - if cli_args.output == None: - base, _ = os.path.split(cli_args.input) - cli_args.output = os.path.join(base, "cravat_converted.txt") - query(cli_args.input, cli_args.output) |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_annotate/cravat_annotate.xml --- a/cravat_annotate/cravat_annotate.xml Wed Jul 18 10:18:43 2018 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 |
| [ |
| @@ -1,25 +0,0 @@ -<tool id="cravat_query" name="CRAVAT Query" version="1.0.0"> - <description>Queries CRAVAT for cancer annotation</description> - <command interpreter="python">cravat_annotate.py -i $input -o $output</command> - - <inputs> - <param format="tabular" name="input" type="data" label="Source file"/> - </inputs> - - <outputs> - <data format="tabular" name="output" /> - </outputs> - - <tests> - <test> - <param name="input" value="input_call.txt"/> - <output name="output" file="Galaxy23-[CRAVAT_Query_on_data_22].tabular"/> - </test> - </tests> - - <help> - This tool queries CRAVAT for cancer annotation. - </help> - -</tool> - |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/__pycache__/base_converter.cpython-36.pyc |
| b |
| Binary file cravat_convert/__pycache__/base_converter.cpython-36.pyc has changed |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/__pycache__/vcf_converter.cpython-36.pyc |
| b |
| Binary file cravat_convert/__pycache__/vcf_converter.cpython-36.pyc has changed |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/base_converter.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_convert/base_converter.py Wed Jul 18 10:25:44 2018 -0400 |
| b |
| @@ -0,0 +1,22 @@ +class BaseConverter(object): + def __init__(self): + self.format_name = None + def check_format(self,*args,**kwargs): + err_msg = 'Converter for %s format has no method check_format' %\ + self.format_name + raise NotImplementedError(err_msg) + def setup(self,*args,**kwargs): + err_msg = 'Converter for %s format has no method setup' %\ + self.format_name + raise NotImplementedError(err_msg) + def convert_line(self,*args,**kwargs): + err_msg = 'Converter for %s format has no method convert_line' %\ + self.format_name + raise NotImplementedError(err_msg) + + +class BadFormatError(Exception): + def __init__(self, message, errors=None): + super(BadFormatError, self).__init__(message) + # Support for custom error codes, if added later + self.errors = errors \ No newline at end of file |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/cravat_convert.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_convert/cravat_convert.py Wed Jul 18 10:25:44 2018 -0400 |
| [ |
| @@ -0,0 +1,80 @@ +import os +import argparse +from vcf_converter import CravatConverter +from __future__ import print_function + +def get_vcf_mapping(): + """ : VCF Headers mapped to their index position in a row of VCF values. + : These are only the mandatory columns, per the VCF spec. + """ + return { + 'CHROM': 0, + 'POS': 1, + 'ID': 2, + 'REF': 3, + 'ALT': 4, + 'QUAL': 5, + 'FILTER': 6, + 'INFO': 7 + } + + +def get_args(): + parser = argparse.ArgumentParser() + parser.add_argument('--input', + '-i', + required = True, + help='Input path to a VCF file for conversion',) + parser.add_argument('--output', + '-o', + default = None, + help = 'Output path to write the cravat file to') + return parser.parse_args() + + +def convert(in_path, out_path=None, cr_sep='\t', cr_newline='\n'): + """ : Convert a VCF file to a Cravat file. + : Arguments: + : in_path: <str> path to input vcf file + : out_path: <str> path to output cravat file. Will defualt to cravat_converted.txt in the input directory. + : cr_sep: <str> the value delimiter for the output cravat file. Default value of '\\t'. + : out_newline: <str> the newline delimiter in the output cravat file. Default of '\\n' + """ + if not out_path: + base, _ = os.path.split(in_path) + out_path = os.path.join(base, "cravat_converted.txt") + + with open(in_path, 'r') as in_file, \ + open(out_path, 'w') as out_file: + + # cr_count will be used to generate the 'TR' field of the cravat rows (first header) + cr_count = 0 + # VCF lines are always assumed to be '+' strand, as VCF doesn't specify that attribute + strand = '+' + # VCF converter. Adjusts position, reference, and alternate for Cravat formatting. + converter = CravatConverter() + # A dictionary of mandatory vcf headers mapped to their row indices + vcf_mapping = get_vcf_mapping() + + for line in in_file: + if line.startswith("#"): + continue + line = line.strip().split() + # row is dict of VCF headers mapped to corresponding values of this line + row = { header: line[index] for header, index in vcf_mapping.items() } + for alt in row["ALT"].split(","): + new_pos, new_ref, new_alt = converter.extract_vcf_variant(strand, row["POS"], row["REF"], alt) + new_pos, new_ref, new_alt = str(new_pos), str(new_ref), str(new_alt) + cr_line = cr_sep.join([ + 'TR' + str(cr_count), row['CHROM'], new_pos, strand, new_ref, new_alt, row['ID'] + ]) + out_file.write(cr_line + cr_newline) + cr_count += 1 + + +if __name__ == "__main__": + cli_args = get_args() + if cli_args.output == None: + base, _ = os.path.split(cli_args.input) + cli_args.output = os.path.join(base, "cravat_converted.txt") + convert(in_path = cli_args.input, out_path = cli_args.output) |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/cravat_convert.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_convert/cravat_convert.xml Wed Jul 18 10:25:44 2018 -0400 |
| b |
| @@ -0,0 +1,20 @@ +<tool id="cravat_convert" name="CRAVAT Convert" version="1.0.0"> + <description>Converts a VCF format file to a Cravat format file</description> + <command interpreter="python">cravat_convert.py -i $input -o $output</command> + + <inputs> + <param format="tabular" name="input" type="data" label="Source file"/> + </inputs> + + <outputs> + <data format="tabular" name="output" /> + </outputs> + + <!-- <tests></tests> --> + + <help> + Converts a VCF format file to a Cravat format file + </help> + +</tool> + |
| b |
| diff -r 326aadaa3dd9 -r a7e641d146f5 cravat_convert/vcf_converter.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_convert/vcf_converter.py Wed Jul 18 10:25:44 2018 -0400 |
| [ |
| b'@@ -0,0 +1,243 @@\n+"""\r\n+A module originally obtained from the cravat package. Modified to use in the vcf\r\n+converter galaxy tool.\r\n+\r\n+\r\n+Register of changes made (Chris Jacoby):\r\n+ 1) Changed imports as galaxy tool won\'t have access to complete cravat python package\r\n+ 2) Defined BadFormatError in BaseConverted file, as I didn\'t have the BadFormatError module\r\n+"""\r\n+\r\n+from base_converter import BaseConverter, BadFormatError\r\n+import re\r\n+\r\n+class CravatConverter(BaseConverter):\r\n+ \r\n+ def __init__(self):\r\n+ self.format_name = \'vcf\'\r\n+ self.samples = []\r\n+ self.var_counter = 0\r\n+ self.addl_cols = [{\'name\':\'phred\',\r\n+ \'title\':\'Phred\',\r\n+ \'type\':\'string\'},\r\n+ {\'name\':\'filter\',\r\n+ \'title\':\'VCF filter\',\r\n+ \'type\':\'string\'},\r\n+ {\'name\':\'zygosity\',\r\n+ \'title\':\'Zygosity\',\r\n+ \'type\':\'string\'},\r\n+ {\'name\':\'alt_reads\',\r\n+ \'title\':\'Alternate reads\',\r\n+ \'type\':\'int\'},\r\n+ {\'name\':\'tot_reads\',\r\n+ \'title\':\'Total reads\',\r\n+ \'type\':\'int\'},\r\n+ {\'name\':\'af\',\r\n+ \'title\':\'Variant allele frequency\',\r\n+ \'type\':\'float\'}]\r\n+ \r\n+ def check_format(self, f): \r\n+ return f.readline().startswith(\'##fileformat=VCF\')\r\n+ \r\n+ def setup(self, f):\r\n+ \r\n+ vcf_line_no = 0\r\n+ for line in f:\r\n+ vcf_line_no += 1\r\n+ if len(line) < 6:\r\n+ continue\r\n+ if line[:6] == \'#CHROM\':\r\n+ toks = re.split(\'\\s+\', line.rstrip())\r\n+ if len(toks) > 8:\r\n+ self.samples = toks[9:]\r\n+ break\r\n+ \r\n+ def convert_line(self, l):\r\n+ if l.startswith(\'#\'): return None\r\n+ self.var_counter += 1\r\n+ toks = l.strip(\'\\r\\n\').split(\'\\t\')\r\n+ all_wdicts = []\r\n+ if len(toks) < 8:\r\n+ raise BadFormatError(\'Wrong VCF format\')\r\n+ [chrom, pos, tag, ref, alts, qual, filter, info] = toks[:8]\r\n+ if tag == \'\':\r\n+ raise BadFormatError(\'ID column is blank\')\r\n+ elif tag == \'.\':\r\n+ tag = \'VAR\' + str(self.var_counter)\r\n+ if chrom[:3] != \'chr\':\r\n+ chrom = \'chr\' + chrom\r\n+ alts = alts.split(\',\')\r\n+ len_alts = len(alts)\r\n+ if len(toks) == 8:\r\n+ for altno in range(len_alts):\r\n+ wdict = None\r\n+ alt = alts[altno]\r\n+ newpos, newref, newalt = self.extract_vcf_variant(\'+\', pos, ref, alt)\r\n+ wdict = {\'tags\':tag,\r\n+ \'chrom\':chrom,\r\n+ \'pos\':newpos,\r\n+ \'ref_base\':newref,\r\n+ \'alt_base\':newalt,\r\n+ \'sample_id\':\'no_sample\',\r\n+ \'phred\': qual,\r\n+ \'filter\': filter}\r\n+ all_wdicts.append(wdict)\r\n+ elif len(toks) > 8:\r\n+ sample_datas = toks[9:]\r\n+ genotype_fields = {}\r\n+ genotype_field_no = 0\r\n+ for genotype_field in toks[8].split(\':\'):\r\n+ genotype_fields[genotype_field] = genotype_field_no\r\n+ genotype_field_no += 1\r\n+ if not (\'GT\' in genotype_fields):\r\n+ raise BadFormatError(\'No GT Field\')\r\n+ gt_field_no = genotype_fields[\'GT\']\r\n+ for sample_no in range(len(sample_datas)):\r\n+ sample = self.samples[sample_no]\r\n+ sample_data = sample_datas[sample_no].split(\':\')\r\n+ gts = {}\r\n+ for gt in sample_data[gt_field_no].replace(\'/\', \'|\').split(\'|\'):\r\n+ '..b'\r\n+ ref_reads = sample_data[genotype_fields[\'AD\']].split(\',\')[0]\r\n+ alt_reads = sample_data[genotype_fields[\'AD\']].split(\',\')[1]\r\n+ elif gt == max(gts.keys()): \r\n+ #if geontype has multiple alt bases, then AD will have #alt1 reads, #alt2 reads\r\n+ alt_reads = sample_data[genotype_fields[\'AD\']].split(\',\')[1]\r\n+ else:\r\n+ alt_reads = sample_data[genotype_fields[\'AD\']].split(\',\')[0] \r\n+ \r\n+ if \'DP\' in genotype_fields and genotype_fields[\'DP\'] <= len(sample_data): \r\n+ depth = sample_data[genotype_fields[\'DP\']] \r\n+ elif alt_reads != \'\' and ref_reads != \'\':\r\n+ #if DP is not present but we have alt and ref reads count, dp = ref+alt\r\n+ depth = int(alt_reads) + int(ref_reads) \r\n+\r\n+ if \'AF\' in genotype_fields and genotype_fields[\'AF\'] <= len(sample_data):\r\n+ af = float(sample_data[genotype_fields[\'AF\']] )\r\n+ elif depth != \'\' and alt_reads != \'\':\r\n+ #if AF not specified, calc it from alt and ref reads\r\n+ af = float(alt_reads) / float(depth)\r\n+ \r\n+ return depth, alt_reads, af\r\n+ \r\n+ def extract_vcf_variant (self, strand, pos, ref, alt):\r\n+\r\n+ reflen = len(ref)\r\n+ altlen = len(alt)\r\n+ \r\n+ # Returns without change if same single nucleotide for ref and alt. \r\n+ if reflen == 1 and altlen == 1 and ref == alt:\r\n+ return pos, ref, alt\r\n+ \r\n+ # Trimming from the start and then the end of the sequence \r\n+ # where the sequences overlap with the same nucleotides\r\n+ new_ref2, new_alt2, new_pos = \\\r\n+ self.trimming_vcf_input(ref, alt, pos, strand)\r\n+ \r\n+ if new_ref2 == \'\':\r\n+ new_ref2 = \'-\'\r\n+ if new_alt2 == \'\':\r\n+ new_alt2 = \'-\'\r\n+ \r\n+ return new_pos, new_ref2, new_alt2\r\n+ \r\n+ # This function looks at the ref and alt sequences and removes \r\n+ # where the overlapping sequences contain the same nucleotide.\r\n+ # This trims from the end first but does not remove the first nucleotide \r\n+ # because based on the format of VCF input the \r\n+ # first nucleotide of the ref and alt sequence occur \r\n+ # at the position specified.\r\n+ # End removed first, not the first nucleotide\r\n+ # Front removed and position changed\r\n+ def trimming_vcf_input(self, ref, alt, pos, strand):\r\n+ pos = int(pos)\r\n+ reflen = len(ref)\r\n+ altlen = len(alt)\r\n+ minlen = min(reflen, altlen)\r\n+ new_ref = ref\r\n+ new_alt = alt\r\n+ new_pos = pos\r\n+ # Trims from the end. Except don\'t remove the first nucleotide. \r\n+ # 1:6530968 CTCA -> GTCTCA becomes C -> GTC.\r\n+ for nt_pos in range(0, minlen - 1): \r\n+ if ref[reflen - nt_pos - 1] == alt[altlen - nt_pos - 1]:\r\n+ new_ref = ref[:reflen - nt_pos - 1]\r\n+ new_alt = alt[:altlen - nt_pos - 1]\r\n+ else:\r\n+ break \r\n+ new_ref_len = len(new_ref)\r\n+ new_alt_len = len(new_alt)\r\n+ minlen = min(new_ref_len, new_alt_len)\r\n+ new_ref2 = new_ref\r\n+ new_alt2 = new_alt\r\n+ # Trims from the start. 1:6530968 G -> GT becomes 1:6530969 - -> T.\r\n+ for nt_pos in range(0, minlen):\r\n+ if new_ref[nt_pos] == new_alt[nt_pos]:\r\n+ if strand == \'+\':\r\n+ new_pos += 1\r\n+ elif strand == \'-\':\r\n+ new_pos -= 1\r\n+ new_ref2 = new_ref[nt_pos + 1:]\r\n+ new_alt2 = new_alt[nt_pos + 1:]\r\n+ else:\r\n+ new_ref2 = new_ref[nt_pos:]\r\n+ new_alt2 = new_alt[nt_pos:]\r\n+ break \r\n+ return new_ref2, new_alt2, new_pos\r\n+\r\n+\r\n+if __name__ == "__main__":\r\n+ c = CravatConverter()\n\\ No newline at end of file\n' |