def prepare_seekr():
parser = argparse.ArgumentParser(description="top level SEEKR program used to prepare and generate all input files necessary for a SEEKR calculation")
parser.add_argument('input_filename', metavar='INPUT_FILENAME', type = str, help="name of SEEKR input file")
args = parser.parse_args() #parse command line arguments
args = vars(args) #converts to dictionary
_get_inputs(args,)
sys_params = _get_sys_params(inputs) #parse general system parameters (structures, forcefield files, directory names from input file)
md_milestones = _parse_milestone_inputs(inputs) #parse milestone CV parameters and milestone values from input file
milestones, md_anchor_list = _generate_milestone_lists(md_milestones) #generate upper/lower bounds of a SINGLE CV for each anchor/Voronoi celli
# TODO generate anchor lists for multiple milestone CV's
# TODO BD milestones
#print(milestones)
_generate_filetree(inputs, sys_params) #creates/clears top level ditectory
filetree_settings = _get_filetree_settings(md_anchor_list)
md_filetree_settings_all = {**filetree_settings, **sys_params}
anchor_dirlist, md_file_paths = filetree.md_filetree(md_filetree_settings_all)
md_settings = _get_md_settings(inputs, md_file_paths) #parse parameters for MD simulations from input file
md_settings_all = {**md_settings, **sys_params, **filetree_settings,}
md.main(md_settings_all, milestones)
print(milestones[-1])
milestone_filename= os.path.join(sys_params['rootdir'], 'milestones.xml')
anchor_list = _group_milestones_to_anchor(milestones, anchor_dirlist, md_file_paths,)
print('Anchor List',anchor_list)
_write_milestone_file(anchor_list, md_settings['master_temperature'],
sys_params['md_time_factor'], sys_params['bd_time_factor'],milestone_filename)
structures = _load_structures(inputs, sys_params)
bd_settings = _get_bd_settings(inputs, sys_params, structures)
bd_milestone = anchor_list[-1]
print(bd_milestone)
#bd_milestone_pair = bd_milestone['%s_pair_list' %bd_milestone['key']][-1]
bd_lower_bound = bd_milestone['milestone_params'][0]['lower_bound']
bd_lower_bound_index = bd_milestone['milestone_params'][0]['lower_milestone_index']
bd_index = bd_milestone['milestone_params'][0]['upper_milestone_index']
b_surf_distance = bd_milestone['milestone_params'][0]['upper_bound']
#bd_milestone_index =
bd_settings.update({'b_surf_distance' : b_surf_distance,
'bd_lower_bound' : bd_lower_bound,
'bd_lower_bound_index' : bd_lower_bound_index,
'bd_index': bd_index,
})
print("BD b surface distance", b_surf_distance)
bd.main(bd_settings)
#!/bin/env python
'''
usage:
genie r1 --fname=FILE [--out=PREFIX] [--nojob] [--dry-run]
options:
--fname=FILE a blank file
--out=PREFIX outname prefix [default: r1]
--nojob no job
--dry-run dry run
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../library')
import md
if __name__ == '__main__':
md.main(docopt(__doc__), ['r1'])
#!/bin/env python
'''ipsych cojo
usage:
ipsych cojo [options] <summary-file>
options:
--out=PREFIX outname prefix [default: ipsych_cojo]
--ldfile=FILE ld reference plink-file [default: ....]
--nsamples=NUMBER number of samples
--nojob if you want to run in front end
'''
from docopt import docopt
import md
if __name__ == '__main__':
md.main(docopt(__doc__))
#!/bin/env
'''
usage:
ldsc munge [options] --sumstats=FILE
options:
--sumstats=FILE daner formatted summary file
--out=PREFIX outname prefix [default: out]
--Nsamples=NUMBER sample size
--other=ARGUMENTS other arguments to pass to ldsc within quotes
--daner if using daner format
--nojob run front end
--dry-run just see the analysis plan but not run
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['munge'])
#!/bin/env python
'''
usage:
gseq bwa [options] --fqlist=FILE
options:
--fqlist=FILE fastq list
--ref-genome=FILE reference fasta file [default: |resources/bwa/reference/b37/human.b37.fasta]
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['bwa'])
#!/bin/env python3
'''
usage:
predixcan simu [options] --gene=FILE
options:
--gene=FILE a file with gene names one per line
--out=PREFIX outname prefix [default: out]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
--int
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['simu'])
usage:
prs prsice [options] --base=FILE --target=FILE
options:
--base=FILE training sample file or file.list
--target=FILE target sample file or file.list
--slower=NUMBER lower limit [default: 0]
--supper=NUMBER upper limit [default: 0.5]
--sinc=NUMBER p-value increments [default: 0.1]
--out=PREFIX outname prefix [default: prs]
--clump if clumping need to be done
--clump_p1=NUMBER clump p1 threshold [default: 1]
--clump_p2=NUMBER clump p2 threshold [default: 1]
--clump_r2=NUMBER clump r2 threshold [default: 0.1]
--clump_kb=NUMBER clump distance [default: 250]
--remove_mhc If you want to remove MHC region
--nojob if should be run in front end
--njobs=NUMBER Number of parallel jobs when running in front end
--dry-run dry run
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['prsice'])
--reml-priors=N1 two numbers
--reml-alg=NUMBER 0(AI) or 1(AI) or 2(EM)
--reml-no-constrain allow negative
--reml-maxit=NUMBER maximum number of iterations
--gxe test gxe
--covar=FILE covariate file
--qcovar=FILE quantitative covariate file
--reml-lrt=NUMBER see GCTA doc
--reml-no-lrt turn off the LRT
--prevalence=NUMBER prevalence of the disease
--dry-run show only the code
--nojob front end
description:
The arguments here are exactly the same arguments used in GCTA software.
To know how to use these arguments, please see the GCTA web page (http://cnsgenomics.com/software/gcta/reml.html).
A typical example submission command for a case-control trait would be like,
genie gcta variance --grm=test --pheno=test.phen --prevalence 0.01 --out test_null
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['variance'])
#!/bin/env python3
'''
usage:
predixcan qc-samples [options] --bfile=FILE --ref=FILE
options:
--bfile=FILE plink file basename or list
--ref=FILE referrence plink file (only basename)
--out=PREFIX outname prefix [default: predixcan]
--skip-sex skip sex checks
--overlap if the ref is within bfile
--cluster=NAME cluster name [default: open]
--nojob run in front end
--dry-run just show the codes
--int control job submission from frontend
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['qc-samples'])
#!/bin/env python3
'''
usage:
predixcan gcta [options] --grm=FILE --pheno=NAME
options:
--grm=BASENAME basename of grm file
--pheno=FILE file or list of plink format phenotype files
--out=PREFIX outname prefix
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
--int
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['gcta'])
#!/bin/env python3
'''
usage: predixcan magma [options] --gwas=FILE --geneannot=FILE --ref=FILE
options:
--gwas=FILE gwas summary statistics
--geneannot=FILE geneannot file
--ref=FILE genotype reference file
--out=PREFIX outname prefix [default: magma]
--nojob run in front end
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs;applicable only when running in front end
--cluster=NAME cluster name [default: open]
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['magma'])
#!/bin/env python3
'''
usage:
mvp gdsforpca [options] --pgen=FILE
options:
--pgen=PREFIX pgen file or .list of pgen files
--nvariants=NUMBER final number of variants [default: 40000]
--out=PREFIX outname prefix [default: predixcan]
--cluster=NAME minerva or genomedk [default: minerva]
--nojob run in front end
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs; applicable only when running in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['gdsforpca'])
#!/bin/env python3
'''
usage:
predixcan qc-dna [options] --vcf=FILE
options:
--vcf=FILE bgzipped and indexed vcf file or list
--ref=FILE bgzipped and indexed referrence vcf file
--ifilter=EXPRESSION filter expression [default: MAF>0.01 & R2>0.8]
--renameINFO
--out=PREFIX outname prefix [default: predixcan]
--cluster=NAME cluster name [default: open]
--nojob run in front end
--dry-run just show the codes
--int control job submission from frontend
--njobs=NUMBER number of parallel jobs; applicable only when running in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['qc-dna'])
#!/bin/python
'''
usage:
sumstats dbsnp [options] --sumstats=FILE
options:
--sumstats=FILE summary statistics file
--out=PREFIX output name prefix [default: out]
--chr=NAME chromosome column name [default: CHR]
--pos=NAME position column name [default: BP]
--ref=NAME ref allele column name [default: A1]
--alt=NAME alternate allele column name [default: A2]
--dry-run dry run
--nojob if need to be run in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['dbsnp'])
#!/bin/python
'''
usage:
sumstats impute [options] --sumstats=FILE --ld=PLINK --N=SAMPLESIZE
options:
--sumstats=FILE munged summary statistics file
--out=PREFIX output name prefix [default: clump_out]
--ld=PLINKFILE plink file without extension
--bed=BEDFILE bed file [default: |resources/sumstats/impute/fizi/locations.bed]
--N=SAMPLESIZE sample size
--nojob if should run in front end
--int
--njobs=NUMBER Number of parallel jobs when running in front end
--dry-run dry run snakemake
--cluster=NAME cluster name [default: open]
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['impute'])
'launches_per_config': inp['fwd_rev_launches_per_config'],
'frame_chunk_size': inp['fwd_rev_frame_chunk_size']
},
'md_file_paths':
md_file_paths, # file paths to the MD directories in the anchor file
'prods_per_anchor': 1, # number of simulations per anchor
#'one_equil_per_anchor':True, # True: all prod simulations will be started from portions of one single equilibration. False: all 50 productions will have their own equilibration
}
md_settings['absolute_mode'] = str(absolute_mode)
md_settings['temp_equil_settings']['temp_range'] = np.concatenate((np.arange(md_settings['temp_equil_settings']['start_temp'],md_settings['temp_equil_settings']['peak_temp'],md_settings['temp_equil_settings']['temp_increment']), \
np.arange(md_settings['temp_equil_settings']['peak_temp'],md_settings['temp_equil_settings']['end_temp'],-md_settings['temp_equil_settings']['temp_increment']), [md_settings['temp_equil_settings']['end_temp']]))
md_settings_all = dict(md_settings.items() + sys_params.items() +
tcl.items())
md.main(md_settings_all)
#print "md_settings:", md_settings
if sys_params['bd']:
bd_receptor_dry_pqr = parser.get_structure(
'bd_receptor_dry_pqr', sys_params['bd_rec_pqr_filename'], pqr=True)
bd_settings={ # settings for the bd model
'rec_struct':bd_receptor_dry_pqr,
'lig_configs': lig_configs,
'temperature':master_temperature,
'threads':int(inp['bd_threads']),
'fhpd_numtraj':inp['fhpd_numtraj'],
# reaction sites information in milestone_settings below
'browndye_bin_dir':inp['browndye_bin_dir'],
#!/bin/env python
'''
usage:
gcta cojo (--cojo-slct | --cojo-joint | --cojo-cond=FILE ) --cojo-file=SUMMARY [options]
options:
--cojo-file=SUMMARY Summary file
--bfile=PLINK Plink file [default: |resources/magma/g1000_eur]
--chr=NUMBER Chromosome number
--maf=NUMBER Minor allele frequency
--out=PREFIX Outname prefix [default: genie_cojo]
--extract=LIST a file with list of SNPs
--cojo-slct Select multiple associated SNPs through a stepwise selection procedure
--cojo-joint Estimate the joint effects of a subset of SNPs
--cojo-cond=FILE Perform single-SNP association analyses conditional on a set of SNPs
--nojob if front end
--dry-run dry run
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['cojo'])
#!/bin/env python3
'''
usage:
predixcan chunkvcf [options] --vcf=FILE
options:
--vcf=VCF vcf file or .list of files
--nvariants=NUMBER number of variants [default: 500000]
--nsamples=NUMBER number of samples [default: 5000]
--out=PREFIX outname prefix [default: predixcan]
--cluster=NAME minerva or genomedk [default: minerva]
--nojob run in front end
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['chunkvcf'])
#!/bin/env python
'''
usage:
gwas genabel [options] --pheno=FILE
options:
--pheno=FILE phenotype file
--covar=FILE covariate file
--geno=FILE genotypefile or .list in genabel format
--out=PREFIX outname prefix [deafult: genabel_out]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['genabel'])
#!/bin/env python
'''
usage:
prs hub [options] --pheno=FILE --covar=FILE
options:
--pheno=FILE phenotype file
--covar=FILE covariate file
--score=FILE score file or .list
--out=PREFIX outname prefix [deafult: hub_out]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['hub'])
#!/bin/env python3
'''
usage:
predixcan wgcna [options] --expr=RDS
options:
--expr=RDS expression rds file
--beta=NUMBER beta value [default: NULL]
--out=PREFIX outname prefix [default: WGCNA]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
--int
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['wgcna'])
#!/bin/env python
'''
usage:
genie hess [options] <sumstat>
options:
--out=PREFIX outname prefix [default: genie_gsmr]
--nojob run on front end
--dry-run dry run
<sumstat> sumstat files. Split by chromosome, filename with * in place of chr number.
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../library')
import md
if __name__ == '__main__':
md.main(docopt(__doc__), ['hess'])
'''
usage:
geneset magma [options] <summary-file>
options:
--out=PREFIX outname prefix[default: output]
--geneloc=FILE geneloc file [default: |resources/magma/NCBI37.3.gene.loc.symbols]
--ldfile=FILE reference file in plink format [default: |resources/magma/g1000_eur]
--ncol=NAME name of the column containing the sample size [default: Neff]
--nsamples=NUMBER alternatively, the number of samples can be assigned like this
--genesets=FILE genesets file in gmt format [default: |resources/magma/gmt.list]
--nojob run on the front end
--dry-run dry run
--njobs=NUMBER number of parallel jobs [default: 1]
<summary-file> a single summary file or a text file(with extension .list) containing the list of summary files
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['magma', 'vegas', 'fastbat'])
'''
usage:
sumstats clump [options] --sumstats=FILE --ld=PLINKFILE
options:
--sumstats=FILE munged summary statistics file
--out=PREFIX output name prefix [default: clump_out]
--ld=PLINKFILE plink file without extension
--p1=NUMBER pvalue threshold 1 [default: 1]
--p2=NUMBER pvalue threshold 2 [default: 1]
--r2=NUMBER rsquared value [default: 0.1]
--distance=NUMBER window distance in kb [default: 250]
--nojob if should run in front end
--int
--njobs=NUMBER Number of parallel jobs when running in front end
--dry-run dry run snakemake
--other=ARGS other arguments to pass to plink with quotes
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
from md import process_list
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['clump'])
#!/bin/env python3
'''
usage:
predixcan train_module [options] --module=FILE --out=NAME
options:
--gds=FILE genotype gds file [default: merged.gds]
--bfile=NAME plink base name [default: merged]
--module=FILE file with list of module genes [default: module.names.list]
--out=PREFIX outname prefix
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
--int
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['train_module'])
#!/bin/env python3
'''
usage:
gwas plink [options] --pheno=FILE --covar=FILE
options:
--pheno=FILE phenotype file
--covar=FILE covariate file
--geno=FILE genotypefile or .list (without plink extension)
--out=PREFIX outname prefix [deafult: genabel_out]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['plink'])
'''
usage:
predixcan train [options] --gds=FILE --expr=FILE
options:
--gds=FILE genotype gds file
--expr=FILE expression file
--genes=FILE a file with list of genes with co-ordinates
--priors=FILE priors file
--grouping=FILE grouping file
--nested
--out=PREFIX outname prefix [default: predixcan]
--cluster=NAME genomedk or minerva [default: minerva]
--nojob run in front end
--dry-run just show the codes
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
--int
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['train'])
#!/bin/env python3
'''
usage:
predixcan predict [options] --gds=FILE --db=FILE
options:
--gds=FILE genotype gds file
--db=FILE prediction model db file
--genes=FILE a file with list of genes with co-ordinates
--out=PREFIX outname prefix [default: predixcan]
--nojob run in front end
--cluster=NAME cluster name [default: minerva]
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments,['predict'])
#!/bin/env python3
'''
usage:
predixcan novel [options] --gwas=FILE
options:
--gwas=FILE gwas summary statistics
--annot=FILE gencode annot [default: |resources/predixcan/gencode.v27.build37.txt]
--out=PREFIX outname prefix [default: predixcan]
--nojob run in front end
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['novel'])
#!/bin/env python3
'''
usage:
predixcan covar [options] --expr=FILE
options:
--expr=FILE expression matrix file
--thold=NUMBER correlation value to filter [default: 0.8]
--cutoff=NUMBER filtering cutoff 20 or 50 percent [default: 0.2]
--out=PREFIX outname prefix [default: predixcan]
--nojob run in front end
--dry-run just show the codes
--int submit jobs from front end
--njobs=NUMBER number of parallel jobs; applicable only when running
in front end
'''
from docopt import docopt
import sys
sys.path.insert(1, sys.path[0] + '/../../../library')
import md
arguments = docopt(__doc__)
if __name__ == '__main__':
md.main(arguments, ['covar'])
