# HG changeset patch # User in_silico # Date 1528815865 14400 # Node ID 152227fa7851e791e9ac8297f95d4db06edeb371 # Parent 3da92d3056714ded9c8ac5ec770cb8dbea22d2df Uploaded diff -r 3da92d305671 -r 152227fa7851 cravat_annotate/cravat_annotate.py --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_annotate/cravat_annotate.py Tue Jun 12 11:04:25 2018 -0400 @@ -0,0 +1,261 @@ +""" +A galaxy wrapper for the /rest/service/query API endpoint on Cravat. + + +Notes on Mapping: +----------------- +The CravatQuery class uses static method 'from_array' to interpret an array of values +into a query string for the /rest/service/query API service on the cravat server. +This involves using a mapping dictionary to know how to associate the array's index positions +in to query-ing attributes, such as the chromosome, position, etc. The CravatQuery +class contains a default value ('default_mapping'); however, this could also be +offered as a user-configurable option. + + +Remaining Items (including possible expansion features): +----------------- +TODO: Possibly provide user-configurability of CravatQuery array index mapping +TODO: Possibly provide user-configurability of delimiter value +TODO: Check if chromosomes are 0 or 1 based indexing +TODO: Port 'write headers' option and include in user prompts in galaxy xml +TODO: Try-catch structure on the query call to cravat so if one bad query doesn't get back a response, + the rest of the run can still execute. Report this to user. +""" + + +import requests +import json +import sys +import re +### +import ipdb + + +class CravatQueryException(Exception): + + def __init__(self, message, errors=None): + super(CravatQueryException, self).__init__(message) + # Support for custom error codes + self.errors = errors + + +class CravatQuery(object): + """ + : A class for handling Cravat query strings. + : Args (all required): + : chr - Chromosome + : pos - Position + : strand - Strand + : ref - Reference Base + : alt - Alternate Base + """ + + # The endpoint that CravatQuerys are submitted to + endpoint = 'http://cravat.us/CRAVAT/rest/service/query' + + # The value delimiter used in the Cravat input file to delimit values + delimiter = "\t" + + # Defualt indices for intepretting a cravat file's row of data in to a CravatQuery + default_mapping = { + 'chromosome': 1, + 'position': 2, + 'strand': 3, + 'reference': 4, + 'alternate': 5 + } + + # Defualt values. Used as backup for CravatQuery to resolve query with incomplete information + default_values = { + 'strand': '+' + } + + # 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. + response_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 __init__(self, _chr, pos, strand, ref, alt): + # '_chr' used to avoid naming confliction with python built-in 'chr' + self.chromosome = CravatQuery.format_chromosome(_chr) + self.position = pos + self.strand = strand + self.reference = ref + self.alternate = alt + self.values = [self.chromosome, self.position, self.strand, self.reference, self.alternate] + + + def __str__(self): + """ + : Represent the CravatQuery as a valid query string for call to Cravat server + """ + return "_".join(map(lambda x: str(x), self.values)) + + + def as_query_string(self): + return str(self) + + + @staticmethod + def from_dictionary(d): + """ + : Instantiate a CravatQuery from a dictionary representation. + : Args: + : d : A dictionary representing a CravatQuery, containing keys: [{}] + """.format(CravatQuery.query_keys) + + for key in CravatQuery.query_keys: + if key not in d: + raise CravatQueryException("CravatQuery.from_dictionary requires keys: [{}], however key: '{}' was not provided " + .format(CravatQuery.query_keys, key)) + return CravatQuery(d["chromosome"], d["position"], d["strand"], d["reference"], d["alternate"]) + + + @staticmethod + def from_array(array, mapping=None): + """ + : Instantiate a CravatQuery from an array of values. Useful when translating read lines from a file. + : Args: + : fmt - Either 'cr' or 'vcf', describing input format + : array - The values to instantiate the CravatQuery from + : mapping - Optional. A dictionary associating cravat parameters to indicies in the array. + Valid values are: 'chromosome', 'position', 'strand', 'reference', 'alternate' + """ + + # Set the mapping value. Either recieved from user, or obtained via defualt associated to 'fmt' + if mapping == None: + mapping = CravatQuery.default_mapping + + # Build a dict of cravat querying keys to values. + d = {} + for key in CravatQuery.query_keys: + # Try to get index position from mapping by the key, and value from array by the index + if key in mapping: + index = mapping[key] + d[key] = array[index] + # If index not provided in mapping, check if there is a defualt value + elif key in CravatQuery.default_values: + d[key] = CravatQuery.default_values[key] + # Unable to get value for querying key, meaning can't construct the minimum requirements for query + else: + raise CravatQueryException("CravatQuery.from_array requires a mapping index for key: '{}', however value was not provided".format(key)) + return CravatQuery.from_dictionary(d) + + + + @staticmethod + def format_chromosome(_chr): + """ + : Format a chromosome for use as query parameter. '_chr' name used to avoid python built-in name confliction. + : Args: + : _chr - Either an interger [1,23], or 'x'/'X', or 'y'/'Y', or a string of the form + : 'chr' where '' is one of the previously described values + """ + inRange = lambda x: 1 <= x and x <= 23 + _chr = _chr.lower() + _chr = _chr.strip('chr') + # Handler interger chromosomes 1 to 23 + try: + _chr = int(_chr) + if inRange(_chr): + return 'chr' + str(_chr) + else: + raise CravatQueryException("Chromsomme of '{}' was out of range [1,23]".format(_chr)) + except: + pass + # Handle chromosomes chromosomes x and y + if _chr == 'x' or _chr == 'y': + return 'chr' + _chr + raise CravatQueryException("Unable to resolve input: '{}' into a valid chromosome representation".format(_chr)) + + + @staticmethod + def jump_header(in_file, out_file, headerlines=0): + """ + : Jumps over a header space of line number 'headerlines'. Sets up in_file so that + : the next execution of in_file.readline() will return the first non-header line. + """ + in_file.seek(0) + for line in range(headerlines): + in_file.readline() + + +def main(in_path, out_path, pre_callback=None, user_mapping=None): + """ + : Read the file line by line and use data to query cravat server. + : Args: + : - fmt : 'cr' or 'vcf'. The input format + : - in_path : Path to input file + : - in_path : Path to output file + : - header_callback : A function to handle the header space. Executed + before main loop. Recieves in_file, out_file, and fmt as argumnets + """ + + with open(in_path, 'r') as in_file, \ + open(out_path, 'w') as out_file: + + # Perform any pre-processing steps, such as jumping a header space + if pre_callback: + pre_callback(in_file, out_file, fmt) + + # main loop + for line in in_file: + + # Create query from line of input data + line = line.strip().split('\t') + query = CravatQuery.from_array(line, user_mapping) + # Make request, and write respone data + call = requests.get(CravatQuery.endpoint, params={ 'mutation': query.as_query_string }) + ipdb.set_trace() + try: + if call.status_code != 200 or call.text == "": + raise CravatQueryException("Bad Server Response. Respone code: '{}', Response Text: '{}'".format(call.status_code, call.text)) + json_response = json.loads(call.text) + wrote = False + for key, val in json_response.items(): + # Set numeric values to uniform format + try: + val = float(val) + val = format(val, ".4f") + except: + pass + if wrote: + out_file.write("\t") + out_file.write(val) + wrote = True + out_file.write("\n") + except CravatQueryException as e: + print(e) + + + + +if __name__ == "__main__": + + # Input and output file paths, obtained form command line + in_path = sys.argv[1] + out_path = sys.argv[2] + + # Possibly allow user mapping configuration thourgh here. Not fully implemented + if len(sys.argv) > 2: + user_mapping = sys.argv[3] + + # Run the main operation + main(in_path, out_path) \ No newline at end of file diff -r 3da92d305671 -r 152227fa7851 cravat_annotate/cravat_annotate.xml --- /dev/null Thu Jan 01 00:00:00 1970 +0000 +++ b/cravat_annotate/cravat_annotate.xml Tue Jun 12 11:04:25 2018 -0400 @@ -0,0 +1,25 @@ + + Queries CRAVAT for cancer annotation + cravat_annotate.py $input $output + + + + + + + + + + + + + + + + + + This tool queries CRAVAT for cancer annotation. + + + + diff -r 3da92d305671 -r 152227fa7851 cravat_annotate_mutations-18ce5c6169ef/cravat_annotate.py --- a/cravat_annotate_mutations-18ce5c6169ef/cravat_annotate.py Tue Jun 12 10:56:54 2018 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,83 +0,0 @@ -import requests -import json -import sys -from __builtin__ import False -#Gets the input and output from galaxy -input_filename = sys.argv[1] -output_filename = sys.argv[2] - -#opens each file, in to read, out to write -in_file = open(input_filename, "r") -out_file = open(output_filename, "w") - - -#sets replacements to replace each space in genomic coordinates with an underscore to run with the query -replacements = {' ':'_'} -#so we only print out the Keys once -write_header = True - -#loops through the input file line by line -for line in in_file: - #strips the input line of \n (new line) and replaces every space with an underscore - line = "_".join( line.split() ) - #gets request from CRAVAT server with the inputed mutation line - call = requests.get('http://staging.cravat.us/CRAVAT/rest/service/query', params={'mutation': line} ) - #puts the string of data into a json dictionary - json_data = json.loads(call.text) - #manually sets the order of the Keys to the same as CRAVAT Server - keys = ["Chromosome","Position","Strand","Reference base(s)","Alternate base(s)","HUGO symbol", - "Sequence ontology transcript","Sequence ontology protein change","Sequence ontology", - "Sequence ontology all transcripts","ExAC total allele frequency", - "ExAC allele frequency (African/African American)","ExAC allele frequency (Latino)", - "ExAC allele frequency (East Asian)","ExAC allele frequency (Finnish)", - "ExAC allele frequency (Non-Finnish European)","ExAC allele frequency (Other)", - "ExAC allele frequency (South Asian)", "1000 Genomes allele frequency", - "ESP6500 allele frequency (European American)","ESP6500 allele frequency (African American)", - "Transcript in COSMIC","Protein sequence change in COSMIC", - "Occurrences in COSMIC [exact nucleotide change]","Mappability Warning","Driver Genes", - "TARGET","dbSNP","MuPIT Link"] - print json_data - - #Spit out first 8 or 9, then loop through rest and print out - for key in json_data: - if key not in keys: - keys.append(key) - - #used so we only print out the Keys once - if write_header == True: - #writes out the keys of the dictionary - out_file.write('\t'.join(keys) + '\n') - write_header = False - - #sets value to the first value in the first key - value = json_data[keys[0]] - #actually writes out the value - out_file.write(value) - #print "key[" + key[0] + "] value[" + str(value) + "]" - #sets value to the second key - value = json_data[keys[1]] - out_file.write('\t' + value) - #print "key[" + key[1] + "] value[" + str(value) + "]" - - - #loops through all other values for each key - for key in keys[2:]: - #strips the value - value = json_data[key].strip() - #another try, except statement to convert the rest of the values to floats, and then round them to four decimals - try: - value = float(value) - value = '%.4f'%value - except: - pass - #writes out the value with a tab after for galaxy formatting - out_file.write("\t" + str(value)) - #print for debugging - print "key[" + key + "] value[" + str(value) + "]" - #creates a new line for the next set of values - out_file.write('\n') - - -#closes both files -in_file.close() -out_file.close() diff -r 3da92d305671 -r 152227fa7851 cravat_annotate_mutations-18ce5c6169ef/cravat_annotate.xml --- a/cravat_annotate_mutations-18ce5c6169ef/cravat_annotate.xml Tue Jun 12 10:56:54 2018 -0400 +++ /dev/null Thu Jan 01 00:00:00 1970 +0000 @@ -1,23 +0,0 @@ - - Queries CRAVAT for cancer annotation - cravat_annotate.py $input $output - - - - - - - - - - - - - - - - This tool queries CRAVAT for cancer annotation. - - - -