-
Notifications
You must be signed in to change notification settings - Fork 5
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
- Loading branch information
1 parent
15231ba
commit 2b27cdc
Showing
30 changed files
with
1,345 additions
and
169 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file not shown.
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,98 @@ | ||
| from __future__ import division | ||
| import sys | ||
| import os | ||
| import errno | ||
| from config_settings import ConfigSectionMap | ||
| from logging_subprocess import * | ||
| from log_modules import * | ||
|
|
||
| """ Sanity Check Methods""" | ||
| def make_sure_path_exists(out_path): | ||
| """ | ||
| Fuction to make sure output folder exists. If not, create it. | ||
| :param: out_path | ||
| :return: null/exception | ||
| """ | ||
| try: | ||
| os.makedirs(out_path) | ||
| except OSError as exception: | ||
| if exception.errno != errno.EEXIST: | ||
| keep_logging('\nErrors in output folder path! please change the output path or analysis name\n', | ||
| '\nErrors in output folder path! please change the output path or analysis name\n', logger, | ||
| 'info') | ||
| exit() | ||
|
|
||
| def make_sure_files_exists(vcf_file_array): | ||
| """ | ||
| Function to make sure the variant call output files exists and are not empty. | ||
| :param: vcf_file_array | ||
| :return: null or exception | ||
| """ | ||
| not_found_files = [] | ||
| for files in vcf_file_array: | ||
| ori_unmapped_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "unmapped.bed_positions") | ||
| ori_proximate_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_final.vcf_no_proximate_snp.vcf_positions_array") | ||
| ori_variant_position_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_final.vcf_no_proximate_snp.vcf") | ||
| ori_5bp_mpileup_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "aln_mpileup_raw.vcf_5bp_indel_removed.vcf") | ||
| ori_mpileup_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "aln_mpileup_raw.vcf") | ||
| ori_filter_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_final.vcf") | ||
| ori_indel_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_indel_final.vcf") | ||
|
|
||
| if os.path.isfile(ori_unmapped_file) and os.path.isfile(ori_proximate_file) and os.path.isfile( | ||
| ori_variant_position_file) and os.path.isfile(ori_5bp_mpileup_file) and os.path.isfile( | ||
| ori_mpileup_file) and os.path.isfile(ori_filter_file) and os.path.isfile(ori_indel_file): | ||
| continue | ||
| else: | ||
| not_found_files.append(files) | ||
| if len(not_found_files) > 0: | ||
| for i in not_found_files: | ||
| keep_logging('Error finding variant calling output files for: %s' % os.path.basename(i.replace('_filter2_final.vcf_no_proximate_snp.vcf', '')), | ||
| 'Error finding variant calling output files for: %s' % os.path.basename(i.replace('_filter2_final.vcf_no_proximate_snp.vcf', '')), logger, 'exception') | ||
| exit() | ||
|
|
||
| def make_sure_label_files_exists(vcf_file_array, uniq_snp_positions, uniq_indel_positions): | ||
| """ | ||
| Function to make sure the variant call output files exists and are not empty. | ||
| :param: vcf_file_array | ||
| :return: null or exception | ||
| """ | ||
| not_found_files = [] | ||
| found_incomplete = [] | ||
| for files in vcf_file_array: | ||
|
|
||
| snps_label_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_final.vcf_no_proximate_snp.vcf_positions_label") | ||
|
|
||
| indel_label_file = files.replace("filter2_final.vcf_no_proximate_snp.vcf", | ||
| "filter2_indel_final.vcf_indel_positions_label") | ||
|
|
||
| num_snps_label_lines = sum(1 for line in open('%s' % snps_label_file)) | ||
| num_indel_label_lines = sum(1 for line in open('%s' % indel_label_file)) | ||
|
|
||
| if os.path.isfile(snps_label_file) and os.path.isfile(indel_label_file): | ||
| if num_snps_label_lines == uniq_snp_positions and num_indel_label_lines == uniq_indel_positions: | ||
| continue | ||
| else: | ||
| found_incomplete.append(files) | ||
| else: | ||
| not_found_files.append(files) | ||
| # keep_logging('Error finding variant calling output files: %s' % files, 'Error finding variant calling output files: %s' % files, logger, 'exception') | ||
| if len(not_found_files) > 0: | ||
| for i in not_found_files: | ||
| keep_logging('Error finding core_prep output files for: %s' % os.path.basename(i.replace('_filter2_final.vcf_no_proximate_snp.vcf', '')), | ||
| 'Error finding core_prep output files for: %s' % os.path.basename(i.replace('_filter2_final.vcf_no_proximate_snp.vcf', '')), logger, 'exception') | ||
| exit() | ||
|
|
||
| if len(found_incomplete) > 0: | ||
| for i in found_incomplete: | ||
| tmp_file = os.path.basename(i.replace('_filter2_final.vcf_no_proximate_snp.vcf', '')) | ||
| keep_logging('core_prep step failed for: %s. Rerun %s.pbs' % (tmp_file, i), | ||
| 'core_prep step failed for: %s. Rerun %s.pbs' % (tmp_file, i), logger, 'exception') | ||
| exit() |
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,24 @@ | ||
| __author__ = 'alipirani' | ||
| import os | ||
| from config_settings import ConfigSectionMap | ||
| from logging_subprocess import * | ||
| from log_modules import * | ||
|
|
||
| def fasttree(tree_dir, input_fasta, cluster, logger, Config): | ||
| keep_logging('Running Fasttree on input: %s' % input_fasta, 'Running Fasttree on input: %s' % input_fasta, logger, 'info') | ||
| fasttree_cmd = "%s/%s/%s -nt %s > %s/%s_FastTree.tree" % (ConfigSectionMap("bin_path", Config)['binbase'], ConfigSectionMap("fasttree", Config)['fasttree_bin'], ConfigSectionMap("fasttree", Config)['base_cmd'], input_fasta, tree_dir, (os.path.basename(input_fasta)).replace('.fa', '')) | ||
| keep_logging('%s' % fasttree_cmd, '%s' % fasttree_cmd, logger, 'info') | ||
| if cluster == "parallel-local" or cluster == "local": | ||
| call("cd %s" % tree_dir, logger) | ||
| call(fasttree_cmd, logger) | ||
| elif cluster == "cluster": | ||
| call("cd %s" % tree_dir, logger) | ||
| call(fasttree_cmd, logger) | ||
| elif cluster == "parallel-cluster": | ||
| job_file_name = "%s/fasttree_%s.pbs" % (tree_dir, os.path.basename(input_fasta)) | ||
| job_name = os.path.basename(job_file_name) | ||
| job_print_string = "#PBS -N %s\n#PBS -M %s\n#PBS -m %s\n#PBS -V\n#PBS -l nodes=1:ppn=4,mem=47000mb,walltime=76:00:00\n#PBS -q %s\n#PBS -A %s\n#PBS -l qos=flux\ncd %s\n%s" % (job_name, ConfigSectionMap("scheduler", Config)['email'], ConfigSectionMap("scheduler", Config)['notification'], ConfigSectionMap("scheduler", Config)['queue'], ConfigSectionMap("scheduler", Config)['flux_account'], tree_dir, fasttree_cmd) | ||
| f1=open(job_file_name, 'w+') | ||
| f1.write(job_print_string) | ||
| f1.close() | ||
| call("qsub %s" % job_file_name, logger) |
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,15 @@ | ||
| __author__ = 'alipirani' | ||
| import os | ||
| from config_settings import ConfigSectionMap | ||
| from logging_subprocess import * | ||
| from log_modules import * | ||
|
|
||
| def gubbins(gubbins_dir, input_fasta, logger, Config): | ||
| keep_logging('\nRunning Gubbins on input: %s\n' % input_fasta, '\nRunning Gubbins on input: %s\n' % input_fasta, logger, | ||
| 'info') | ||
| call("cd %s" % ConfigSectionMap("gubbins", Config)['gubbins_bin'], logger) | ||
| gubbins_cmd = "%s/%s --prefix %s/%s %s" % (ConfigSectionMap("gubbins", Config)['gubbins_bin'], ConfigSectionMap("gubbins", Config)['base_cmd'], gubbins_dir, (os.path.basename(input_fasta)).replace('.fa', ''), input_fasta) | ||
| #call(gubbins_cmd, logger) | ||
| keep_logging('\nRunning Gubbins: %s' % input_fasta, '\nRunning Gubbins: %s\n' % input_fasta, | ||
| logger, | ||
| 'info') |
Binary file not shown.
Binary file not shown.
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file not shown.
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| Original file line number | Diff line number | Diff line change |
|---|---|---|
| @@ -0,0 +1,26 @@ | ||
| __author__ = 'alipirani' | ||
| import os | ||
| from config_settings import ConfigSectionMap | ||
| from logging_subprocess import * | ||
| from log_modules import * | ||
|
|
||
|
|
||
| def raxml(tree_dir, input_fasta, jobrun, logger, Config): | ||
| keep_logging('Running RAXML on input: %s' % input_fasta, 'Running RAXML on input: %s' % input_fasta, logger, 'info') | ||
| raxml_cmd = "%s/%s/%s %s -s %s -n %s_raxML" % (ConfigSectionMap("bin_path", Config)['binbase'], ConfigSectionMap("raxml", Config)['raxml_bin'], ConfigSectionMap("raxml", Config)['base_cmd'], ConfigSectionMap("raxml", Config)['parameters'], input_fasta, (os.path.basename(input_fasta)).replace('.fa', '')) | ||
| keep_logging('%s' % raxml_cmd, '%s' % raxml_cmd, logger, 'info') | ||
| if jobrun == "parallel-local" or jobrun == "local": | ||
| call("cd %s" % tree_dir, logger) | ||
| call(raxml_cmd, logger) | ||
| elif jobrun == "cluster": | ||
| call("cd %s" % tree_dir, logger) | ||
| call(raxml_cmd, logger) | ||
| elif jobrun == "parallel-cluster": | ||
| job_file_name = "%s/raxml_%s.pbs" % (tree_dir, os.path.basename(input_fasta)) | ||
| job_name = os.path.basename(job_file_name) | ||
| job_print_string = "#PBS -N %s\n#PBS -M %s\n#PBS -m %s\n#PBS -V\n#PBS -l nodes=1:ppn=4,mem=47000mb,walltime=76:00:00\n#PBS -q %s\n#PBS -A %s\n#PBS -l qos=flux\ncd %s\n%s" % (job_name, ConfigSectionMap("scheduler", Config)['email'], ConfigSectionMap("scheduler", Config)['notification'], ConfigSectionMap("scheduler", Config)['queue'], ConfigSectionMap("scheduler", Config)['flux_account'], tree_dir, raxml_cmd) | ||
| f1=open(job_file_name, 'w+') | ||
| f1.write(job_print_string) | ||
| f1.close() | ||
| #os.system("qsub %s" % job_file_name) | ||
| call("qsub %s" % job_file_name, logger) |
Binary file not shown.
Binary file not shown.
Binary file not shown.
Oops, something went wrong.