def signalp_task(path_orfs,out_dir,tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.signalp'.format(out_dir,out_name)]
cmd = '{0!s} -f short -n {1!s} {2!s}'.format(fg.tool_path_check(TOOLS_DICT['signalp'].full_exe[0]),trgs[-1],path_orfs)
name = 'signalp_' + out_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def salmon_unpaired_task(index,unpaired,out_name,gene_map,out_dir,cpu_cap,tasks):
trgs = []
cmd = '{0!s} quant -i {1!s} -l U -r {2!s} -o {3!s}/{4!s} --geneMap {5!s} -p {6!s} --extraSensitive'.format(
fg.tool_path_check(TOOLS_DICT['salmon'].full_exe[0]),index,unpaired,out_dir,out_name,gene_map,cpu_cap)
name = 'salmon_unpaired_' + os.path.basename(index)
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def signalp_task(opc, path_orfs, out_dir, tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.signalp'.format(out_dir, out_name)]
cmd = '{0!s} -f short -n {1!s} {2!s}'.format(tool_path_check(TOOLS_DICT['signalp'].full_exe[0]), trgs[-1], path_orfs)
name = 'signalp_' + out_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err)
def prinseq_task(opc, out_dir, input_1, input_2, basename, opts, tasks):
''' Defines prinseq task. Uses GEN_PATH_DIR(), PATH_PRINSEQ
Params :
input_1 - a list of 1/left fastq files
input_2 - a list of 2/right fastq files
basename - the basename for all output files
opts - optional params for trinity task.
tasks = the tasks that this task is dependent on
'''
trgs = [
'{0!s}/{1!s}_1_{2!s}'.format(out_dir, basename,
os.path.basename(input_1)),
'{0!s}/{1!s}_2_{2!s}'.format(out_dir, basename,
os.path.basename(input_2))
]
pseudo_trgs = [
'{0!s}/{1!s}_{2!s}.fastq'.format(out_dir, basename, x)
for x in range(1, 3)
]
cmd = (
'perl {0!s} -fastq {1!s} -fastq2 {2!s} --out_format 3 --out_good {3!s}/{4!s} '
'--out_bad null --trim_qual_left 20 --trim_qual_right 20 --trim_qual_type min '
'--min_len 55 --trim_tail_left 8 --trim_tail_right 8 {5!s} -log; mv {6!s} {7!s};'
' mv {8!s} {9!s};').format(
tool_path_check(TOOLS_DICT['prinseq'].full_exe[0]), input_1,
input_2, out_dir, basename, opts, pseudo_trgs[0], trgs[0],
pseudo_trgs[1], trgs[1])
name = basename
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
name=name,
stdout=out,
stderr=err,
targets=trgs)
def salmon_unpaired_task(opc, index, unpaired, out_name, gene_map, out_dir,
cpu_cap, tasks):
trgs = [
'{0!s}/{1!s}_quant.sf'.format(out_dir, out_name),
'{0!s}/{1!s}_quant.genes.sf'.format(out_dir, out_name)
]
if len(gene_map) > 0:
trans_gene_map = ' --geneMap {0!s}'.format(gene_map)
cmd = (
'{0!s} quant -i {1!s} -l U -r {2!s} -o {3!s}/{4!s} {5!s} '
#'-p {6!s} --extraSensitive; cp {3!s}/{4!s}/quant.sf '
'-p {6!s} --dumpEq --extraSensitive; cp {3!s}/{4!s}/quant.sf '
'{3!s}/{4!s}_quant.sf; cp {3!s}/{4!s}/quant.genes.sf '
'{3!s}/{4!s}_quant.genes.sf').format(
tool_path_check(TOOLS_DICT['salmon'].full_exe[0]), index, unpaired,
out_dir, out_name, trans_gene_map, cpu_cap)
name = 'salmon_unpaired_' + os.path.basename(
index) + '_' + os.path.basename(unpaired)
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err,
cpu=cpu_cap)
def kallisto_task(index,out_dir,out_name,left,right,tasks):
trgs = []
cmd = '{0!s} quant -i {1!s} -o {2!s}/{3!s} {4!s} {5!s}'.format(
fg.tool_path_check(TOOLS_DICT['kallisto'].full_exe[0]),index,out_dir,out_name,left,right)
name = 'kallisto'
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def busco_task(opc, dbs, assembly_path, assembly_name, out_dir, reference_name,
cpu_cap, tasks):
''' Defines the busco task. Uses PATH_DIR, PATH_BUSCO, PATH_BUSCO_REFERENCE
Params :
reference_name - Name of the reference file to be used by busco
cpu_cap - the cpu limit to be gicen to busco.
tasks - a list of tasks that this task is dependant on.
'''
trgs = [
'{0!s}/run_busco_{1!s}_{2!s}'.format(out_dir, assembly_name,
reference_name)
]
busco_db = dbs['busco_' + reference_name]
cmd = ('cd {0!s}; python {1!s} '
'-o busco_{3!s}_{2!s} -i {4!s} -l {5!s}/{2!s}_odb9 -m '
'tran -f -c {6!s}').format(
out_dir, tool_path_check(TOOLS_DICT['busco'].full_exe[0]),
reference_name, assembly_name, assembly_path,
busco_db.call_path, cpu_cap)
name = 'busco_' + reference_name + '_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
cpu=cpu_cap,
stdout=out,
stderr=err)
def trimmomatic_task(opc, out_dir, left, right, cpu_cap, basename, tasks):
base_str = '{0!s}/{1!s}'.format(out_dir, basename)
trgs = [
base_str + '_1_' + os.path.basename(left),
base_str + '_2_' + os.path.basename(right)
]
orphans = [
base_str + '_1s_' + os.path.basename(left),
base_str + '_2s_' + os.path.basename(right)
]
cmd = ('java -jar {0!s} PE -threads {3!s} {1!s} {2!s} {5!s} {4!s} {7!s} '
'{6!s} ILLUMINACLIP:{8!s}:2:30:10 LEADING:3 TRAILING:3 '
'SLIDINGWINDOW:4:15 MINLEN:35').format(
tool_path_check(TOOLS_DICT['trimmomatic'].full_exe[0]), left,
right, cpu_cap, orphans[0], trgs[0], orphans[1], trgs[1],
TOOLS_DICT['trimmomatic'].full_exe[1])
name = basename
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
name=name,
stdout=out,
stderr=err,
targets=trgs,
cpu=cpu_cap)
def build_salmon_task(path_assembly,assembly_name,out_dir,cpu_cap,tasks):
trgs = ['{0!s}/{1!s}_salmon'.format(out_dir, assembly_name)]
cmd = '{0!s} index --transcripts {1!s} --index {2!s}/{3!s}_salmon --threads {4!s} --type quasi'.format(
fg.tool_path_check(TOOLS_DICT['salmon'].full_exe[0]),path_assembly, out_dir, assembly_name, cpu_cap)
name = 'build_salmon_' + assembly_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err, cpu=cpu_cap)
def gene_trans_map_task(path_assembly,out_dir,tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
trgs = ['{0!s}/{1!s}.gene_trans_map'.format(out_dir, assembly_name)]
cmd = '{0!s} {1!s} > {2!s}'.format(fg.tool_path_check(TOOLS_DICT['trinity'].full_exe[1]),path_assembly,trgs[0])
name = 'gene_trans_map_' + assembly_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def tmhmm_task(path_orfs,out_dir,tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.tmhmm'.format(out_dir, out_name)]
cmd = 'cd {0!s}; {1!s} --short < {2!s} > {3!s}'.format(out_dir,fg.tool_path_check(TOOLS_DICT['tmhmm'].full_exe[0]),path_orfs,trgs[0])
name = 'tmhmm_' + out_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def ret():
for t in other_dependencies:
try:
if( not d.finished()):
return False
except Task.ExitCodeException:
return False
assembly_files = sorted(os.listdir(reads_dir))
assembly_files = [os.path.join(reads_dir, f) for f in assembly_files]
new_lefts = [[g for g in assembly_files if(os.path.basename(f) in g)] for f in lefts]
new_lefts = [k[0] for k in new_lefts if(len(k) > 0)]
new_rights = [[g for g in assembly_files if(os.path.basename(f) in g)] for f in rights]
new_rights = [k[0] for k in new_rights if(len(k) > 0)]
new_singles = [[g for g in assembly_files if(os.path.basename(f) in g)] for f in singles]
new_singles = [k[0] for k in new_singles if(len(k) > 0)]
if(len(new_lefts) == len(lefts) and len(new_rights) == len(rights) and len(new_singles) == len(singles)
and len(new_lefts)+len(new_singles) != 0):
new_lefts = ','.join(new_lefts+new_singles)
new_rights = ','.join(new_rights)
new_lefts = '--left '+new_lefts if(len(new_lefts) > 0) else ''
new_rights = '--right '+new_rights if(len(new_rights) > 0) else ''
cmd = '{0!s} --assembly {1!s} {2!s} {3!s} --threads {4!s} {5!s} --output {6!s}'.format(
fg.tool_path_check(TOOLS_DICT['transrate'].full_exe[0]), assembly_path, new_lefts,
new_rights, cpu_cap, reference, transrate_dir)
transrate_task.command = cmd
else:
print('Unable to match input files with trimmed output. Continuing transrate using input files instead.')
return True
def build_kallisto_task(assembly_path, assembly_name,out_dir,tasks):
trgs = []
cmd = '{0!s} index -i {1!s}/{2!s}_kallisto {3!s}'.format(
fg.tool_path_check(TOOLS_DICT['kallisto'].full_exe[0]),out_dir,assembly_name,assembly_path)
name = 'build_kallisto'
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def salmon_task(opc, index, left, right, out_name, gene_map, out_dir, cpu_cap,
tasks):
trgs = [
'{0!s}/{1!s}/quant.sf'.format(out_dir, out_name),
'{0!s}/{1!s}/quant.genes.sf'.format(out_dir, out_name)
]
if len(gene_map) > 0:
trans_gene_map = ' --geneMap {0!s}'.format(gene_map)
cmd = (
'{0!s} quant -i {1!s} -l IU -1 {2!s} -2 {3!s} -o {4!s}/{5!s} '
#'--geneMap {6!s} -p {7!s} --extraSensitive; cp {4!s}/{5!s}/quant.sf '
'{6!s} -p {7!s} --dumpEq'
).format(
#cmd = '{0!s} quant -i {1!s} -l IU -1 {2!s} -2 {3!s} -o {4!s}/{5!s}
#--geneMap {6!s} -p {7!s} --extraSensitive --numBootstraps 30 --biasCorrect ; cp ' \
tool_path_check(TOOLS_DICT['salmon'].full_exe[0]),
index,
left,
right,
out_dir,
out_name,
trans_gene_map,
cpu_cap)
name = os.path.basename(index) + '_' + os.path.basename(left)
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err,
cpu=cpu_cap)
def trinity_task(opc, path_assembly, out_dir, fastq, fastq2, unpaired,
cpu_cap_trin, cpu_cap_bfly, mem_trin, mem_bfly,
normalize_flag, tasks):
''' Defines the trinity task. Uses GEN_PATH_DIR(), PATH_TRINITY, NAME_ASSEMBLY
Params :
left - a 1/left fastq files
right - a 2/right fastq files
cpu_cap - number of threads used by trinity
tasks - a list of tasks that this task is dependent on
'''
normalize_flag = '--normalize_reads' if (normalize_flag) else ''
input_str = ''
if (unpaired != [] and fastq == []):
input_str += '--single ' + ','.join(unpaired)
if (fastq != []):
input_str += '--left ' + ','.join(fastq + unpaired)
input_str += ' --right ' + ','.join(fastq2)
trgs = [path_assembly]
cmd = (
'{0!s} --seqType fq {1!s} --CPU {2!s} --max_memory {3!s}G --bflyCalculateCPU {4!s} '
'--output {6!s}/trinity; cp {6!s}/trinity/Trinity.fasta {7!s};'
).format(tool_path_check(TOOLS_DICT['trinity'].full_exe[0]), input_str,
cpu_cap_trin, mem_trin, normalize_flag, mem_bfly, out_dir,
trgs[0])
name = 'trinity_assembly'
out, err = gen_logs(opc.path_logs, name)
cpu_cap = max(cpu_cap_trin, cpu_cap_bfly)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
cpu=cpu_cap,
stdout=out,
stderr=err)
def pfam_seq_task(path_orfs,out_dir, cpu_cap, tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.pfam_tblout'.format(out_dir,out_name)]
cmd = '{0!s} --cpu {1!s} --tblout {2!s} {3!s} {4!s}'.format(
fg.tool_path_check(TOOLS_DICT['hmmer'].full_exe[0]),cpu_cap,trgs[0],PATH_PFAM_DATABASE, path_orfs)
name = 'pfam_tblout_' + out_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def salmon_task(index,left,right,out_name,gene_map,out_dir,cpu_cap,tasks):
trgs = ['{0!s}/{1!s}_quant.sf'.format(out_dir,out_name),'{0!s}/{1!s}_quant.genes.sf'.format(out_dir,out_name)]
cmd = '{0!s} quant -i {1!s} -l IU -1 {2!s} -2 {3!s} -o {4!s}/{5!s} --geneMap {6!s} -p {7!s} --extraSensitive; cp ' \
'{4!s}/{5!s}/quant.sf {4!s}/{5!s}_quant.sf; cp {4!s}/{5!s}/quant.genes.sf {4!s}/{5!s}_quant.genes.sf'.format(
fg.tool_path_check(TOOLS_DICT['salmon'].full_exe[0]),index,left,right,out_dir,out_name,gene_map,cpu_cap)
name = 'salmon_' + os.path.basename(index) + '_' + os.path.basename(left)
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def tmhmm_task(opc, path_orfs, out_dir, tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.tmhmm'.format(out_dir, out_name)]
cmd = 'cd {0!s}; {1!s} --short < {2!s} > {3!s}'.format(
out_dir, tool_path_check(TOOLS_DICT['tmhmm'].full_exe[0]),
path_orfs, trgs[0])
name = 'tmhmm_' + out_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err)
def pfam_seq_task(opc, dbs, path_orfs, out_dir, cpu_cap, tasks):
out_name = os.path.basename(path_orfs).split('.')[0]
trgs = ['{0!s}/{1!s}.pfam_tblout'.format(out_dir, out_name)]
cmd = '{0!s} --cpu {1!s} --tblout {2!s} {3!s} {4!s}'.format(
tool_path_check(TOOLS_DICT['hmmer'].full_exe[0]), cpu_cap,
trgs[0], dbs['pfam'].call_path, path_orfs)
name = 'pfam_tblout_' + out_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err, cpu=cpu_cap)
def gene_trans_map_task(opc, path_assembly, out_dir, tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
trgs = ['{0!s}/{1!s}.gene_trans_map'.format(out_dir, assembly_name)]
cmd = 'perl {0!s} {1!s} > {2!s}'.format(
tool_path_check(TOOLS_DICT['trinity'].full_exe[1]),
path_assembly, trgs[0])
name = 'gene_trans_map_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err)
def build_blast_task(path_db,out_dir,dbtype,tasks,log_flag=True):
trgs = []
#title doesn't seem to change the out name .. it's still xx.gz.psq, etc? CHECK.
title = os.path.basename(path_assembly).split('.')[0]
cmd = 'gunzip -c {0!s} | {1!s} -in - -dbtype {2!s} -title {3!s} -out {4!s}'.format(
path_assembly, fg.tool_path_check(TOOLS_DICT['blast'].full_exe[0]),dbtype,title,out_dir)
name = 'build_blastplus_db_' + title
out, err = fg.GEN_LOGS(name) if(log_flag) else (None, None)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def intersect_bed_task(out_dir,bam_file,bed_reference,output_name,tasks):
'''
'''
trgs = ['{0!s}/{1!s}.bed'.format(out_dir,output_name)]
cmd = '{0!s} intersect -abam {1!s} -b {2!s} -wb -bed > {3!s}'.format(
fg.tool_path_check(TOOLS_DICT['bedtools'].full_exe[0]),bam_file,bed_reference,trgs[0])
name = 'intersect_bed_'+output_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def build_bowtie_task(path_assembly, assembly_name, out_dir, tasks):
'''
'''
trgs = ['{0!s}/{1!s}.1.bt2'.format(out_dir,assembly_name)]
cmd = '{0!s} --offrate 1 -f {1!s} {2!s}/{3!s}'.format(
fg.tool_path_check(TOOLS_DICT['bowtie2'].full_exe[0]), path_assembly, out_dir, assembly_name)
name = 'build_bowtie_' + assembly_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def express_task(assembly_path,out_dir,out_name,bam_input,tasks):
'''
'''
trgs = ['{0!s}/{1!s}.xprs'.format(out_dir,out_name)]
cmd = ('mkdir {1!s}/{2!s}; {0!s} --output-dir {1!s}/{2!s} {3!s} {4!s}; mv '
'{1!s}/{2!s}/results.xprs {5!s}; rm -rf {1!s}/{2!s};').format(
fg.tool_path_check(TOOLS_DICT['express'].full_exe[0]),out_dir,out_name,assembly_path,bam_input,trgs[0])
name = 'express_'+out_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def rnammer_task(path_assembly, out_dir, tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
path_to_rnammer = os.path.dirname(TOOLS_DICT['rnammer'].folder_name)
trgs = ['{0!s}/{1!s}.fasta.rnammer.gff'.format(out_dir,assembly_name)]
cmd = ("cd {0!s}; {1!s} --transcriptome {2!s} --path_to_rnammer {4!s} "
"--org_type euk; cd -").format(out_dir,fg.tool_path_check(TOOLS_DICT['rnammer'].full_exe[0]),
path_assembly,path_to_rnammer)
name = 'rnammer_' + assembly_name
out,err = GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def transdecoder_longorfs_task(path_assembly, path_transdecoder_output, cpu_cap, tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
longorf_outbase = os.path.join(path_transdecoder_output, assembly_name + '.fasta.transdecoder_dir')
trgs = ['{0!s}/longest_orfs.pep'.format(longorf_outbase),'{0!s}/longest_orfs.gff3'.format(longorf_outbase),
'{0!s}/longest_orfs.cds'.format(longorf_outbase)]
cmd = ("mkdir -p {0!s}; cd {0!s}; {1!s} -t {2!s}").format(path_transdecoder_output,
fg.tool_path_check(TOOLS_DICT['transdecoder'].full_exe[0]),path_assembly,cpu_cap)
name = 'TransDecoder_LongORFs_' + assembly_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def rnammer_task(opc, path_assembly, out_dir, tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
path_to_rnammer = os.path.dirname(TOOLS_DICT['rnammer'].folder_name)
trgs = ['{0!s}/{1!s}.fasta.rnammer.gff'.format(out_dir, assembly_name)]
cmd = ("cd {0!s}; {1!s} --transcriptome {2!s} --path_to_rnammer {4!s} "
"--org_type euk; cd -").format(
out_dir, tool_path_check(TOOLS_DICT['rnammer'].full_exe[0]),
path_assembly, path_to_rnammer)
name = 'rnammer_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err)
def trimmomatic_unpaired_task(out_dir,input1, cpu_cap, basename, tasks):
form = lambda s, i : s.format(out_dir, basename, os.path.basename(i))
trgs = [form('{0!s}/{1!s}_{2!s}', input1)]
orphans = [form('{0!s}/{1!s}_orphans_{2!s}', input1)]
cmd = ('java -jar {0!s} SE -threads {4!s} {1!s} {2!s} {3!s} ILLUMINACLIP:'
'{5!s}:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:35'
).format(fg.tool_path_check(TOOLS_DICT['trimmomatic'].full_exe[0]), input1, trgs[0], orphans[0],cpu_cap,
TOOLS_DICT['trimmomatic'].full_exe[2]) #PATH_TRIMMOMATIC_ADAPTERS_SINGLE)
name = basename
out, err = fg.GEN_LOGS(name)
return Task(command=cmd, dependencies=tasks, name=name, stdout=out, stderr=err, targets=trgs, cpu=cpu_cap)
def transdecoder_longorfs_task(opc, path_assembly, path_transdecoder_output, cpu_cap, tasks):
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
longorf_outbase = os.path.join(path_transdecoder_output, opc.assembly_name + '.fasta.transdecoder_dir')
trgs = ['{0!s}/longest_orfs.pep'.format(longorf_outbase),
'{0!s}/longest_orfs.gff3'.format(longorf_outbase),
'{0!s}/longest_orfs.cds'.format(longorf_outbase)]
cmd = ("mkdir -p {0!s}; cd {0!s}; {1!s} -t {2!s}").format(path_transdecoder_output,
tool_path_check(TOOLS_DICT['transdecoder'].full_exe[0]), path_assembly, cpu_cap)
name = 'TransDecoder_LongORFs_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err, cpu=cpu_cap)
def bowtie2_task(bowtie2_index,out_dir,fastq1,fastq2,out_name,opt,cpu_cap,tasks):
'''
'''
opts = ['-a -t --end-to-end', '-t --local']
trgs = ['{0!s}/{1!s}.bam'.format(out_dir,out_name)]
cmd = ('{0!s} {1!s} -L {2!s} -N 1 --maxins 800 --threads {3!s} -x {4!s} -1 '
'{5!s} -2 {6!s} | samtools view -Sb - > {7!s} ').format(fg.tool_path_check(TOOLS_DICT['bowtie2'].full_exe[1]),
opts[opt],22,cpu_cap,bowtie2_index,fastq1,fastq2,trgs[0])
name = 'bowtie2_'+out_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def prinseq_unpaired_task(out_dir,input1, basename, opts, tasks):
'''
'''
trgs = ['{0!s}/{1!s}_{2!s}'.format(out_dir, basename, os.path.basename(input1))]
cmd = ('perl {0!s} -fastq {1!s} --out_format 3 --out_good {2!s}/{3!s} --out_bad null '
'--trim_qual_left 20 --trim_qual_right 20 --trim_qual_type min --min_len 35 '
'--trim_tail_left 8 --trim_tail_right 8 {4!s} -log; mv {2!s}/{3!s}.fastq {5!s}'
).format(fg.tool_path_check(TOOLS_DICT['prinseq'].full_exe[0]), input1, out_dir,
basename, opts, trgs[0])
name = basename
out, err = fg.GEN_LOGS(name)
return Task(command = cmd, dependencies=tasks, name=name, stdout=out, stderr=err, targets=trgs)
def build_bowtie_task(opc, path_assembly, assembly_name, out_dir, tasks):
trgs = ['{0!s}/{1!s}.1.bt2'.format(out_dir, assembly_name)]
cmd = '{0!s} --offrate 1 -f {1!s} {2!s}/{3!s}'.format(
tool_path_check(TOOLS_DICT['bowtie2'].full_exe[0]), path_assembly,
out_dir, assembly_name)
name = 'build_bowtie_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def cegma_task(out_dir,assembly,cpu_cap, tasks):
''' Defines the cegma task. Uses PATH_DIR, PATH_CEGMA, NAME_ASSEMBLY.
Params :
cpu_cap - number of threads to be used by cegma
tasks - a list of tasks that this task is dependant on (trinity_task)
'''
assembly_name = os.path.basename(assembly).split('.fa')[0]
trgs = ['{0!s}/{1!s}.completeness_report'.format(out_dir,assembly_name)]
cmd = '{0!s} -g {1!s} -v -o {3!s}/{2!s} -T {4!s}'.format(fg.tool_path_check(TOOLS_DICT['cegma'].full_exe[0]),
assembly,assembly_name,out_dir,cpu_cap)
name = 'cegma'
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,cpu=cpu_cap,stdout=out,stderr=err)
def diamond_task(blast_type, out_dir, path_query, ref, cpu_cap, tasks):
''' valid blast_types: "blastx", "blastp" '''
base_ref = os.path.basename(ref)
query_name = os.path.basename(path_query).split('.')[0]
trgs = ['{0!s}/{1!s}_{2!s}.diamond_{3!s}'.format(out_dir, query_name, base_ref, blast_type)]
pseudo_trgs = ['{0!s}/diamond_{1!s}_{2!s}'.format(out_dir, base_ref, blast_type)]
cmd = ('{0!s} {1!s} --db {2!s} --query {3!s} --daa {4!s} --tmpdir {5!s} '
'--max-target-seqs 20 --sensitive --threads {6!s} --evalue 0.001; {0!s} view '
'--daa {4!s}.daa --out {7!s};').format(
fg.tool_path_check(TOOLS_DICT['diamond'].full_exe[0]), blast_type, ref, path_query, pseudo_trgs[0], out_dir,
cpu_cap, trgs[0])
name = 'diamond_{0!s}_{1!s}_{2!s}'.format(blast_type, base_ref, query_name)
out, err = fg.GEN_LOGS(name)
return Task(command=cmd, dependencies=tasks, cpu=cpu_cap, targets=trgs, name=name, stdout=out, stderr=err)
def trimmomatic_task(out_dir,left, right, cpu_cap, basename, tasks):
form = lambda s, i : s.format(out_dir, basename, os.path.basename(i))
trgs = [form('{0!s}/{1!s}_1_{2!s}', left),
form('{0!s}/{1!s}_2_{2!s}', right)]
orphans = [form('{0!s}/{1!s}_1s_{2!s}', left),
form('{0!s}/{1!s}_2s_{2!s}', right)]
cmd = ('java -jar {0!s} PE -threads {3!s} {1!s} {2!s} {5!s} {4!s} {7!s} '
'{6!s} ILLUMINACLIP:{8!s}:2:30:10 LEADING:3 TRAILING:3 '
'SLIDINGWINDOW:4:15 MINLEN:35').format(
fg.tool_path_check(TOOLS_DICT['trimmomatic'].full_exe[0]), left, right, cpu_cap, orphans[0], trgs[0],
orphans[1], trgs[1], TOOLS_DICT['trimmomatic'].full_exe[1]) #PATH_TRIMMOMATIC_ADAPTERS_PAIRED)
name = basename
out, err = fg.GEN_LOGS(name)
return Task(command=cmd, dependencies=tasks, name=name, stdout=out, stderr=err, targets=trgs, cpu=cpu_cap)
def diamond_task(opc, blast_type, out_dir, path_query, ref, cpu_cap, tasks):
''' valid blast_types: "blastx", "blastp" '''
base_ref = os.path.basename(ref)
query_name = os.path.basename(path_query).split('.')[0]
trgs = ['{0!s}/{1!s}_{2!s}.diamond_{3!s}'.format(out_dir, query_name, base_ref, blast_type)]
pseudo_trgs = ['{0!s}/diamond_{1!s}_{2!s}'.format(out_dir, base_ref, blast_type)]
cmd = ('{0!s} {1!s} --db {2!s} --query {3!s} --daa {4!s} --tmpdir {5!s} '
'--max-target-seqs 20 --sensitive --threads {6!s} --evalue 0.001; {0!s} view '
'--daa {4!s}.daa --out {7!s};').format(
tool_path_check(TOOLS_DICT['diamond'].full_exe[0]), blast_type, ref, path_query,
pseudo_trgs[0], out_dir, cpu_cap, trgs[0])
name = 'diamond_{0!s}_{1!s}_{2!s}'.format(blast_type, base_ref, query_name)
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, cpu=cpu_cap, targets=trgs, name=name, stdout=out, stderr=err)
def kallisto_task(opc, index, out_dir, out_name, left, right, tasks):
# NO TARGETS
trgs = []
cmd = '{0!s} quant -i {1!s} -o {2!s}/{3!s} {4!s} {5!s}'.format(
tool_path_check(TOOLS_DICT['kallisto'].full_exe[0]), index, out_dir,
out_name, left, right)
name = 'kallisto'
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def build_kallisto_task(opc, assembly_path, assembly_name, out_dir, tasks):
# NO TARGETS
trgs = []
cmd = '{0!s} index -i {1!s}/{2!s}_kallisto {3!s}'.format(
tool_path_check(TOOLS_DICT['kallisto'].full_exe[0]), out_dir,
assembly_name, assembly_path)
name = 'build_kallisto'
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def build_blast_task(path_db, out_dir, dbtype, tasks):
trgs = []
# title doesn't seem to change the out name .. it's still xx.gz.psq, etc? CHECK.
title = os.path.basename(path_db).split('.')[0]
cmd = 'gunzip -c {0!s} | {1!s} -in - -dbtype {2!s} -title {3!s} -out {4!s}'.format(
path_db, tool_path_check(TOOLS_DICT['blast'].full_exe[0]), dbtype,
title, out_dir)
name = 'build_blastplus_db_' + title
out, err = gen_db_logs(name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def express_task(opc, bowtie2_index, assembly_path, out_dir, out_name,
bam_input, tasks):
trgs = ['{0!s}/{1!s}.xprs'.format(out_dir, out_name)]
cmd = ('mkdir {1!s}/{2!s}; {0!s} --output-dir {1!s}/{2!s} {3!s} {4!s}; mv '
'{1!s}/{2!s}/results.xprs {5!s}; rm -rf {1!s}/{2!s};').format(
tool_path_check(TOOLS_DICT['express'].full_exe[0]), out_dir,
out_name, assembly_path, bam_input, trgs[0])
name = 'express_' + os.path.basename(bowtie2_index) + '_' + out_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def busco_task(assembly_path, assembly_name, out_dir,reference_name, cpu_cap, tasks):
''' Defines the busco task. Uses PATH_DIR, PATH_BUSCO, PATH_BUSCO_REFERENCE
Params :
reference_name - Name of the reference file to be used by busco
cpu_cap - the cpu limit to be gicen to busco.
tasks - a list of tasks that this task is dependant on.
'''
trgs = ['{0!s}/run_busco_{1!s}_{2!s}'.format(out_dir,assembly_name,reference_name)]
cmd = ('cd {0!s}; /matta1/biotools/anaconda/envs/py3k/bin/python {1!s} '
'-o busco_{3!s}_{2!s} -in {4!s} -l {5!s}/{2!s}_buscos/{2!s} -m trans -f -c {6!s}'
).format(out_dir,fg.tool_path_check(TOOLS_DICT['busco_plant'].full_exe[0]),reference_name,assembly_name,assembly_path,
PATH_BUSCO_REFERENCE,cpu_cap)
name = 'busco_'+ reference_name + '_' + assembly_name
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,cpu=cpu_cap,stdout=out,stderr=err)
def rnaspades_task(path_assembly, out_dir, left, right, unpaired, cpu_cap, tasks):
'''
'''
virtual_target = '{0!s}/rna_spades_out_dir'.format(out_dir)
trgs = [path_assembly]
input_strings = []
if(left!=[]):
input_strings.append('-1 '+left[0])
input_strings.append('-2 '+right[0])
if(unpaired!=[]):
input_strings.append('-s '+unpaired[0])
cmd = '{0!s} {1!s} --threads {2!s} -o {3!s}; cp {3!s}/contigs.fasta {4!s};'.format(
fg.tool_path_check(TOOLS_DICT['rnaspades'].full_exe[0]),' '.join(input_strings),cpu_cap,virtual_target,trgs[0])
name = 'rnaSPAdes_assembly'
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err,cpu=cpu_cap)
def intersect_bed_task(opc, out_dir, bam_file, bed_reference, output_name,
tasks):
trgs = ['{0!s}/{1!s}.bed'.format(out_dir, output_name)]
# cmd = '{0!s} intersect -abam {1!s} -b {2!s} -wb -bed > {3!s}'.format(
cmd = '{0!s} -abam {1!s} -b {2!s} -wb -bed > {3!s}'.format(
tool_path_check(TOOLS_DICT['bedtools'].full_exe[0]), bam_file,
bed_reference, trgs[0])
name = 'intersect_bed_' + os.path.basename(
bed_reference) + '_' + output_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def build_salmon_task(opc, path_assembly, assembly_name, out_dir, cpu_cap,
tasks):
trgs = ['{0!s}/{1!s}_salmon'.format(out_dir, assembly_name)]
cmd = ('{0!s} index --transcripts {1!s} --index {2!s}/{3!s}_salmon '
'--threads {4!s} --type quasi').format(
tool_path_check(TOOLS_DICT['salmon'].full_exe[0]),
path_assembly, out_dir, assembly_name, cpu_cap)
name = 'build_salmon_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err,
cpu=cpu_cap)
def build_diamond_task(path_db_fasta, out_path, tasks):
''' Is there a reason that we aren't checking for the installation of daimond?
'''
title = os.path.basename(out_path)
trgs = ['{0!s}'.format(out_path + '.dmnd')]
cmd = '{0!s} makedb --in {1!s} --db {2!s}'.format(
tool_path_check(TOOLS_DICT['diamond'].full_exe[0]), path_db_fasta,
out_path)
name = 'build_diamond_' + title
out, err = gen_db_logs(name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def blast_task(blast_type, out_dir, path_query, path_db, cpu_cap, tasks):
exe_index = 1
if blast_type == 'blastx':
exe_index = 1
elif blast_type == 'blastp':
exe_index = 2
assembly_name = os.path.basename(path_query).split('.')[0]
db_name = os.path.basename(path_db).split('.')[0]
trgs = ["{0!s}/{1!s}_{2!s}.{3!s}".format(out_dir, assembly_name, db_name, blast_type)]
cmd = ('{0!s} -query {1!s} -db {2!s} -num_threads {3!s} -max_target_seqs 1 '
'-outfmt "6 qseqid sseqid pident length mismatch gapopen qstart qend sstart '
'send evalue bitscore stitle slen" -evalue 0.0001 > {4!s}'
).format(fg.tool_path_check(TOOLS_DICT['blast'].full_exe[exe_index]), path_query, path_db, cpu_cap, trgs[0])
name = '{0!s}_{1!s}_{2!s}'.format(assembly_name, blast_type, db_name)
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,cpu=cpu_cap,stdout=out,stderr=err)
def blast_task(opc, blast_type, out_dir, path_query, path_db, cpu_cap, tasks):
exe_index = 1
if(blast_type == 'blastx'):
exe_index = 1
elif(blast_type == 'blastp'):
exe_index = 2
assembly_name = os.path.basename(path_query).split('.')[0]
db_name = os.path.basename(path_db).split('.')[0]
trgs = ["{0!s}/{1!s}_{2!s}.{3!s}".format(out_dir, assembly_name, db_name, blast_type)]
cmd = ('{0!s} -query {1!s} -db {2!s} -num_threads {3!s} -max_target_seqs 1 '
'-outfmt "6 qseqid sseqid pident length mismatch gapopen qstart qend sstart '
'send evalue bitscore stitle slen" -evalue 0.0001 > {4!s}'
).format(tool_path_check(TOOLS_DICT['blast'].full_exe[exe_index]),
path_query, path_db, cpu_cap, trgs[0])
name = '{0!s}_{1!s}_{2!s}'.format(assembly_name, blast_type, db_name)
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, cpu=cpu_cap, stdout=out, stderr=err)
def bowtie2_unpaired_task(opc, bowtie2_index, out_dir, fastq, out_name, opt,
cpu_cap, tasks):
opts = ['-a -t --end-to-end', '-t --local', '-k 200 --end-to-end']
trgs = ['{0!s}/{1!s}.bam'.format(out_dir, out_name)]
cmd = ('{0!s} {1!s} -L {2!s} -N 1 --threads {3!s} -x {4!s} -U '
'{5!s} | samtools view -Sb - > {6!s} ').format(
tool_path_check(TOOLS_DICT['bowtie2'].full_exe[1]), opts[opt],
22, cpu_cap, bowtie2_index, fastq, trgs[0])
name = 'bowtie2_' + os.path.basename(bowtie2_index) + '_' + out_name
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err,
cpu=cpu_cap)
def pfam_build_task(source, out_root_path, tasks):
''' Trgs seem to be declared without respect to input.
'''
trgs = [
os.path.join(statics.PATH_PFAM_DIR, os.path.basename(source)) + '.h3f'
]
cmd = 'cd {0!s} ; {1!s} -f {2!s};'.format(
statics.PATH_DATABASES,
tool_path_check(TOOLS_DICT['hmmer'].full_exe[1]), source)
name = 'hmmpress_' + os.path.basename(source)
out, err = gen_db_logs(name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
stdout=out,
stderr=err)
def rcorrector_task(opc, out_dir, left, right, cpu_cap, basename, tasks):
trgs = [
'{0!s}/{1!s}.cor.fq'.format(out_dir, os.path.basename(left)),
'{0!s}/{1!s}.corr.fq'.format(out_dir, os.path.basename(right))
]
cmd = 'perl {0!s} -1 {1!s} -2 {2!s} -t {3!s}'.format(
tool_path_check(TOOLS_DICT['rcorrector'].full_exe[0]), left, right,
cpu_cap)
name = 'Rcorrector_' + basename
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
name=name,
stdout=out,
stderr=err,
targets=trgs,
cpu=cpu_cap)
def trimmomatic_unpaired_task(opc, out_dir, input1, cpu_cap, basename, tasks):
trgs = [
'{0!s}/{1!s}_{2!s}'.format(out_dir, basename, os.path.basename(input1))
]
cmd = ('java -jar {0!s} SE -threads {3!s} {1!s} {2!s} ILLUMINACLIP:'
'{4!s}:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:35'
).format(tool_path_check(TOOLS_DICT['trimmomatic'].full_exe[0]),
input1, trgs[0], cpu_cap, TOOLS_DICT['trimmomatic'].
full_exe[2]) # PATH_TRIMMOMATIC_ADAPTERS_SINGLE)
name = basename
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
name=name,
stdout=out,
stderr=err,
targets=trgs,
cpu=cpu_cap)
def prinseq_unpaired_task(opc, out_dir, input1, basename, opts, tasks):
trgs = [
'{0!s}/{1!s}_{2!s}'.format(out_dir, basename, os.path.basename(input1))
]
cmd = (
'perl {0!s} -fastq {1!s} --out_format 3 --out_good {2!s}/{3!s} --out_bad null '
'--trim_qual_left 20 --trim_qual_right 20 --trim_qual_type min --min_len 35 '
'--trim_tail_left 8 --trim_tail_right 8 {4!s} -log; mv {2!s}/{3!s}.fastq {5!s}'
).format(tool_path_check(TOOLS_DICT['prinseq'].full_exe[0]), input1,
out_dir, basename, opts, trgs[0])
name = basename
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
name=name,
stdout=out,
stderr=err,
targets=trgs)
def transrate_task(reads_dir, assembly_path, assembly_name,lefts, rights, singles, out_dir, transrate_dir, cpu_cap, tasks, reference = ''): #, cpu_cap, tasks):
trgs = ['{0!s}/assemblies.csv'.format(transrate_dir),'{0!s}/{1!s}/good.{1!s}.fasta'.format(transrate_dir,assembly_name),'{0!s}/{1!s}/{1!s}.fasta_quant.sf'.format(transrate_dir,assembly_name)]
orig_lefts = lefts
orig_rights = rights
orig_singles = singles
lefts = ','.join(lefts+singles)
rights = ','.join(rights)
lefts = '--left '+lefts if(len(lefts) > 0) else ''
rights = '--right '+rights if(len(rights) > 0) else ''
reference = '--reference ' + reference if(reference != '') else ''
cmd = '{0!s} --assembly {1!s} {2!s} {3!s} --threads {4!s} {5!s} --output {6!s}'.format(
fg.tool_path_check(TOOLS_DICT['transrate'].full_exe[0]), assembly_path, lefts,
rights, cpu_cap, reference, transrate_dir)
name = 'transrate_' + assembly_name
out, err = fg.GEN_LOGS(name)
temp_task = Task(command=cmd, dependencies=[], targets=trgs, name=name, cpu=cpu_cap, stdout=out, stderr=err, max_wall_time=720)
deps = transrate_dep_generator(reads_dir, temp_task, orig_lefts, orig_rights, orig_singles, reference, assembly_path, cpu_cap, transrate_dir, tasks)
temp_task.dependencies = [deps]
return temp_task
def transdecoder_predict_orfs_task(path_assembly,path_transdecoder_output,tasks,pfam_input='',blastp_input=''):
''' Use transdecoder to predict ORF's from input fasta file.
Required for downstream blastp, pfam, tmhmm, signalp.
Targets: *transdecoder.*
'''
pfam,blastp,retain_blastp, retain_pfam = '','','',''
if len(pfam_input) > 0:
pfam = '--retain_pfam_hits ' + pfam_input
retain_pfam = '_retain_pfam'
if len(blastp_input) > 0:
blastp = '--retain_blastp_hits ' + blastp_input
retain_blastp = '_retain_blastp'
assembly_name = os.path.basename(path_assembly).split('.fa')[0]
trgs = ['{0!s}/{1!s}.fasta.transdecoder.pep'.format(path_transdecoder_output,assembly_name),'{0!s}/{1!s}.fasta.transdecoder.bed'.format(path_transdecoder_output,assembly_name)]
cmd = ("mkdir -p {0!s}; cd {0!s}; {1!s} -t {2!s} {3!s} {4!s}").format(path_transdecoder_output,
fg.tool_path_check(TOOLS_DICT['transdecoder'].full_exe[1]),path_assembly,pfam, blastp)
name = 'TransDecoder_Predict' + assembly_name + retain_pfam + retain_blastp
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,stdout=out,stderr=err)
def transrate_task(opc,
reads_dir,
assembly_path,
assembly_name,
lefts,
rights,
out_dir,
transrate_dir,
cpu_cap,
tasks,
reference=''): #, cpu_cap, tasks):
trgs = [
'{0!s}/assemblies.csv'.format(transrate_dir),
'{0!s}/{1!s}/good.{1!s}.fasta'.format(transrate_dir, assembly_name),
'{0!s}/{1!s}/{1!s}.fasta_quant.sf'.format(transrate_dir, assembly_name)
]
orig_lefts = lefts
orig_rights = rights
lefts = ','.join(lefts)
rights = ','.join(rights)
lefts = '--left ' + lefts if (len(lefts) > 0) else ''
rights = '--right ' + rights if (len(rights) > 0) else ''
reference = '--reference ' + reference if (reference != '') else ''
#cmd = '{0!s} --assembly {1!s} --threads {4!s} {5!s} --output {6!s}'.format( #no reads
cmd = '{0!s} --assembly {1!s} {2!s} {3!s} --threads {4!s} {5!s} --output {6!s}'.format(
tool_path_check(TOOLS_DICT['transrate'].full_exe[0]), assembly_path,
lefts, rights, cpu_cap, reference, transrate_dir)
name = 'transrate_' + assembly_name
out, err = gen_logs(opc.path_logs, name)
temp_task = Task(command=cmd,
dependencies=[],
targets=trgs,
name=name,
cpu=cpu_cap,
stdout=out,
stderr=err,
max_wall_time=7200000)
deps = transrate_dep_generator(reads_dir, temp_task, orig_lefts,
orig_rights, reference, assembly_path,
cpu_cap, transrate_dir, tasks)
temp_task.dependencies = [deps]
return temp_task
def prinseq_task(out_dir,input_1, input_2, basename, opts, tasks):
''' Defines prinseq task. Uses GEN_PATH_DIR(), PATH_PRINSEQ
Params :
input_1 - a list of 1/left fastq files
input_2 - a list of 2/right fastq files
basename - the basename for all output files
opts - optional params for trinity task.
tasks = the tasks that this task is dependent on
'''
trgs = ['{0!s}/{1!s}_1_{2!s}'.format(out_dir,basename,os.path.basename(input_1)),
'{0!s}/{1!s}_2_{2!s}'.format(out_dir,basename,os.path.basename(input_2))]
pseudo_trgs = ['{0!s}/{1!s}_{2!s}.fastq'.format(out_dir,basename,x) for x in range(1,3)]
cmd = ('perl {0!s} -fastq {1!s} -fastq2 {2!s} --out_format 3 --out_good {3!s}/{4!s} '
'--out_bad null --trim_qual_left 20 --trim_qual_right 20 --trim_qual_type min '
'--min_len 55 --trim_tail_left 8 --trim_tail_right 8 {5!s} -log; mv {6!s} {7!s};'
' mv {8!s} {9!s};').format(fg.tool_path_check(TOOLS_DICT['prinseq'].full_exe[0]), input_1,
input_2, out_dir, basename, opts,pseudo_trgs[0],trgs[0],pseudo_trgs[1],trgs[1])
name = basename
out,err = fg.GEN_LOGS(name)
return Task(command = cmd, dependencies=tasks, name=name, stdout=out, stderr=err, targets=trgs)
def cegma_task(opc, out_dir, assembly, cpu_cap, tasks):
''' Defines the cegma task. Uses PATH_DIR, PATH_CEGMA, NAME_ASSEMBLY.
Params :
cpu_cap - number of threads to be used by cegma
tasks - a list of tasks that this task is dependant on (trinity_task)
'''
assembly_name = os.path.basename(assembly).split('.fa')[0]
trgs = ['{0!s}/{1!s}.completeness_report'.format(out_dir, assembly_name)]
cmd = '{0!s} -g {1!s} -v -o {3!s}/{2!s} -T {4!s}'.format(
tool_path_check(TOOLS_DICT['cegma'].full_exe[0]), assembly,
assembly_name, out_dir, cpu_cap)
name = 'cegma'
out, err = gen_logs(opc.path_logs, name)
return Task(command=cmd,
dependencies=tasks,
targets=trgs,
name=name,
cpu=cpu_cap,
stdout=out,
stderr=err)
def ret():
for t in other_dependencies:
try:
if (not t.finished()):
return False
except Task.ExitCodeException:
return False
assembly_files = sorted(os.listdir(reads_dir))
assembly_files = [os.path.join(reads_dir, f) for f in assembly_files]
new_lefts = [[g for g in assembly_files if (os.path.basename(f) in g)]
for f in lefts]
new_lefts = [k[0] for k in new_lefts if (len(k) > 0)]
new_rights = [[
g for g in assembly_files if (os.path.basename(f) in g)
] for f in rights]
new_rights = [k[0] for k in new_rights if (len(k) > 0)]
# new_singles = [[g for g in assembly_files if(os.path.basename(f) in g)] for f in singles]
# new_singles = [k[0] for k in new_singles if(len(k) > 0)]
if (len(new_lefts) == len(lefts) and len(new_rights)
== len(rights)): # and len(new_singles) == len(singles)
# and len(new_lefts)+len(new_singles) != 0):
new_lefts = ','.join(new_lefts) # +new_singles)
new_rights = ','.join(new_rights)
new_lefts = '--left ' + new_lefts if (len(new_lefts) > 0) else ''
new_rights = '--right ' + new_rights if (
len(new_rights) > 0) else ''
cmd = '{0!s} --assembly {1!s} {2!s} {3!s} --threads {4!s} {5!s} --output {6!s}'.format(
#cmd = '{0!s} --assembly {1!s} --threads {4!s} {5!s} --output {6!s}'.format( #no reads
tool_path_check(TOOLS_DICT['transrate'].full_exe[0]),
assembly_path,
new_lefts,
new_rights,
cpu_cap,
reference,
transrate_dir)
transrate_task.command = cmd
else:
warnings.warn(
'Unable to match input files with trimmed output. Continuing transrate using input files instead.'
)
return True
def trinity_task(path_assembly, out_dir, fastq, fastq2, unpaired, cpu_cap_trin, cpu_cap_bfly, mem_trin, mem_bfly, normalize_flag, tasks):
''' Defines the trinity task. Uses GEN_PATH_DIR(), PATH_TRINITY, NAME_ASSEMBLY
Params :
left - a 1/left fastq files
right - a 2/right fastq files
cpu_cap - number of threads used by trinity
tasks - a list of tasks that this task is dependent on
'''
normalize_flag = '--normalize_reads' if(normalize_flag) else ''
input_str = ''
if(unpaired!=[] and fastq==[]):
input_str+='--single '+','.join(unpaired)
if(fastq!=[]):
input_str+='--left '+','.join(fastq+unpaired)
input_str+=' --right '+','.join(fastq2)
trgs = [path_assembly]
cmd = ('{0!s} --seqType fq {1!s} --CPU {2!s} --max_memory {3!s}G --bflyCalculateCPU {4!s} '
'--output {6!s}/trinity; cp {6!s}/trinity/Trinity.fasta {7!s};'
).format(fg.tool_path_check(TOOLS_DICT['trinity'].full_exe[0]), input_str, cpu_cap_trin, mem_trin, normalize_flag,
mem_bfly, out_dir, trgs[0])
name = 'trinity_assembly'
out,err = fg.GEN_LOGS(name)
return Task(command=cmd,dependencies=tasks,targets=trgs,name=name,cpu=max(cpu_cap_trin,cpu_cap_bfly),stdout=out,stderr=err)
def pfam_build_task(source, tasks, log_flag=True):
trgs = [PATH_PFAM_DATABASE+'.h3f']
cmd = 'cd {0!s} ; {1!s} -f {2!s};'.format(fg.PATH_DATABASES, fg.tool_path_check(TOOLS_DICT['hmmer'].full_exe[1]), source)
name = 'hmmpress'
out, err = fg.GEN_LOGS(name) if(log_flag) else (None, None)
return Task(command=cmd, dependencies=tasks, targets=trgs, name=name, stdout=out, stderr=err)
