def _annotate_gtf(self, read_length):
annotate_gtf = Job(name='annotate_gtf')
annotate_gtf.invoke('all', self._state_update % 'Generating annotation FASTA files')
prefix = self._get_index_hash(read_length)
# Inputs
gtf = File('%s.gtf' % self._species.name)
chromosomes = self._species.chromosomes
for i in chromosomes:
chr_i = File('%s/chr%s.fa' % (self._species.name, i))
# Uses
annotate_gtf.uses(chr_i, link=Link.INPUT)
# Outputs
features = File('h%s/FEATURES.fa' % prefix)
chrs = File('h%s/GENOME.fa' % prefix)
splices = File('h%s/SPLICES.fa' % prefix)
genes = File('h%s/GENE.fa' % prefix)
# Arguments
annotate_gtf.addArguments(gtf, '-c', self._species.name, '-p h%s/' % prefix, '-l %d' % read_length)
# Uses
annotate_gtf.uses(gtf, link=Link.INPUT)
annotate_gtf.uses(features, link=Link.OUTPUT, transfer=False, register=False)
annotate_gtf.uses(chrs, link=Link.OUTPUT, transfer=False, register=False)
annotate_gtf.uses(splices, link=Link.OUTPUT, transfer=False, register=False)
annotate_gtf.uses(genes, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(annotate_gtf)
def _analyze(self):
analyze = Job(name='analyze_samfile')
analyze.invoke('all', self._state_update % 'Analyzing SAM file')
# Input files
sam_file = File('%s.sam' % self._prefix)
# Output files
genes_counts = File('%s.gene.cnts' % self._prefix)
features_counts = File('%s.feature.cnts' % self._prefix)
ambiguous_genes_counts = File('%s.ambiguousGenes.cnts' % self._prefix)
overlap_genes_counts = File('%s.overlapGene.cnts' % self._prefix)
summary_out = File('%s.summary.out' % self._prefix)
# Arguments
analyze.addArguments(sam_file, '--prefix', self._prefix)
# Uses
analyze.uses(sam_file, link=Link.INPUT)
analyze.uses(genes_counts, link=Link.OUTPUT, transfer=True, register=False)
analyze.uses(features_counts, link=Link.OUTPUT, transfer=True, register=False)
analyze.uses(ambiguous_genes_counts, link=Link.OUTPUT, transfer=True, register=False)
analyze.uses(overlap_genes_counts, link=Link.OUTPUT, transfer=True, register=False)
analyze.uses(summary_out, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(analyze)
def _pre_filter_fastq(self, index, suffix_len):
pre_filter = Job(name='pre_filter_fastq.py')
pre_filter.invoke('all', self._state_update % 'Pre-filter reads file part %d' % (index + 1))
prefix = 'reads%d' % index
# Inputs
reads = File(('x%0' + str(suffix_len) + 'd') % index)
# Outputs
full_fastq = File('%s_full.fastq' % prefix)
reject = File('%s_reject.fastq' % prefix)
stats = File('%s.stats' % prefix)
# Arguments
trims = ','.join([str(i) for i in self._trims])
trims = '0' if trims == ',' else trims
pre_filter.addArguments(reads, '-r', '%d' % self._read_length, '-t', '%s' % trims)
pre_filter.addArguments('-p', prefix)
# Uses
pre_filter.uses(reads, link=Link.INPUT)
for t in self._trims:
fastq_t = File('%s_%d.fastq' % (prefix, t))
pre_filter.uses(fastq_t, link=Link.OUTPUT, transfer=False, register=False)
pre_filter.uses(full_fastq, link=Link.OUTPUT, transfer=False, register=False)
pre_filter.uses(reject, link=Link.OUTPUT, transfer=False, register=False)
pre_filter.uses(stats, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(pre_filter)
def _perm_index(self, index_type, read_length, read_format='fastq', seed='F2'):
perm_index = Job(name='perm')
perm_index.invoke('all', self._state_update % 'Pre-computing %s index file' % index_type.capitalize())
prefix = self._get_index_hash(read_length)
# Input files
fa_input = File('h%s/%s.fa' % (prefix, index_type))
# Output files
hash_v = self._get_index_hash(read_length, seed)
index = File('h%d_%s_%s_%s.index' % (hash_v, index_type, seed, read_length))
# Arguments
perm_index.addArguments(fa_input, '%d' % read_length, '--readFormat', read_format, '--seed', seed)
perm_index.addArguments('-s', index)
# Uses
perm_index.uses(fa_input, link=Link.INPUT)
# Save this file
perm_index.uses(index, link=Link.OUTPUT, transfer=True, register=True)
self.adag.addJob(perm_index)
return perm_index
def _bar_plot(self):
bar_plot = Job(name='bar_plot')
bar_plot.invoke('all', self._state_update % 'Plot summary out file')
# Input files
summary_file = File('%s.summary.out' % self._prefix)
# Output files
pdf_file = File('%s.ps' % self._prefix)
# Arguments
bar_plot.addArguments('--output-file', pdf_file, summary_file)
# Uses
bar_plot.uses(summary_file, link=Link.INPUT)
bar_plot.uses(pdf_file, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(bar_plot)
def _farish_compact(self):
farish_compact = Job(name='farish_compact')
farish_compact.invoke('all', self._state_update % 'Farish Compact')
# Input files
unmapped = File('%s.unmapped.fastq' % self._prefix)
# Output files
compact = File('%s.compact' % self._prefix)
# Arguments
farish_compact.addArguments(unmapped, '-o', compact)
# Uses
farish_compact.uses(unmapped, link=Link.INPUT)
farish_compact.uses(compact, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(farish_compact)
def _clipr(self, clip_to, reads, tag):
anchor = self._compute_clip_seed(self._read_length)
clip_reads = Job(name='clipR')
clip_reads.invoke('all', self._state_update % 'Generate new splice candidates')
seed = 'F%s' % self._clip_seed
mismatches = self._clip_mismatches
# Input files
prefix = self._get_index_hash(self._read_length)
fa = File('h%s/%s.fa' % (prefix, clip_to.upper()))
reads_txt = File('%s_%s_reads.txt' % (tag, clip_to.lower()))
for i in self._range():
# Input files
reads_i = File(reads % i)
# Output files
file_type = 'sam'
path, file_name, ext = GTFAR._get_filename_parts(reads_i.name)
sam_mapping = '%s_A_%d_%d_%d_%s.%s' % (clip_to.upper(), self._clip_seed, mismatches, anchor, file_name, file_type)
fastq_out = File('%s_miss_%s%s' % (file_name, clip_to, ext))
# Uses
clip_reads.uses(reads_i, link=Link.INPUT)
clip_reads.uses(fastq_out, link=Link.OUTPUT, transfer=False, register=False)
clip_reads.uses(sam_mapping, link=Link.OUTPUT, transfer=False, register=False)
# Output files
log = File('%s_%s.log' % (tag, clip_to.lower()))
# Arguments
clip_reads.addArguments(fa, reads_txt, '--seed %s' % seed, '--anchorL %d' % anchor, '-e', '-v %d' % mismatches)
clip_reads.addArguments('-s', '-u', '--noSamHeader', '--ignoreDummyR %d' % 40, '--ignoreRepeatR %d' % 15)
clip_reads.setStdout(log)
# Uses
clip_reads.uses(fa, link=Link.INPUT)
clip_reads.uses(reads_txt, link=Link.INPUT)
clip_reads.uses(log, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(clip_reads)
def _perm(self, index_type, map_to, reads, tag, output_sam=False):
perm = Job(name='perm')
perm.invoke('all', self._state_update % 'Map reads to %s' % map_to.capitalize())
# Input files
hash_v = self._get_index_hash(self._read_length, 'F%d' % self._seed)
index = File('h%d_%s_F%d_%d.index' % (hash_v, map_to, self._seed, self._read_length))
reads_txt = File('%s_%s_reads.txt' % (tag, map_to.lower()))
for i in self._range():
# Input files
reads_i = File(reads % i)
# Output files
file_type = 'sam' if output_sam else 'mapping'
path, file_name, ext = GTFAR._get_filename_parts(reads_i.name)
sam_mapping = '%s_B_%d_%d_%s.%s' % (map_to.upper(), self._seed, self._mismatches, file_name, file_type)
fastq_out = File('%s_miss_%s%s' % (file_name, map_to, ext))
# Uses
perm.uses(reads_i, link=Link.INPUT)
perm.uses(fastq_out, link=Link.OUTPUT, transfer=False, register=False)
perm.uses(sam_mapping, link=Link.OUTPUT, transfer=False, register=False)
# Output files
log = File('%s_%s.log' % (tag, map_to.upper()))
# Arguments
perm.addArguments(index, reads_txt, '--seed F%d' % self._seed, '-v %d' % self._mismatches, '-B', '--printNM')
perm.addArguments('-u', '-s', '-T %d' % self._read_length)
if output_sam:
perm.addArguments('--noSamHeader', '--outputFormat', 'sam')
perm.setStdout(log)
# Uses
perm.uses(index, link=Link.INPUT)
perm.uses(reads_txt, link=Link.INPUT)
perm.uses(log, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(perm)
def _fastq_split(self, splits=2, suffix_len=2):
fastq_split = Job(name='fastq-split')
fastq_split.invoke('all', self._state_update % 'Splitting input reads file into %d parts' % splits)
# Inputs
reads = File(self._reads)
# Arguments
fastq_split.addArguments(reads, '%d' % splits)
# Uses
fastq_split.uses(reads, link=Link.INPUT)
for i in range(splits):
split_i = File(('x%0' + str(suffix_len) + 'd') % i)
# Outputs
fastq_split.uses(split_i, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(fastq_split)
def _transcript_prediction(self):
transcript_prediction = Job(name='transcript_prediction')
transcript_prediction.invoke('all', self._state_update % 'Transcript Prediction')
# Input files
features_counts = File('%s.feature.cnts' % self._prefix)
gtf = File('%s.splice_candidates.gtf' % self._prefix)
# Output files
transcript_counts = File('%s.transcripts.cnts' % self._prefix)
# Arguments
transcript_prediction.addArguments(features_counts, '-g', gtf)
# Uses
transcript_prediction.setStdout(transcript_counts)
transcript_prediction.uses(features_counts, link=Link.INPUT)
transcript_prediction.uses(gtf, link=Link.INPUT)
transcript_prediction.uses(transcript_counts, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(transcript_prediction)
def _parse_clipped_alignment(self, input_file):
parse_clipped_alignment = Job(name='parse_clipped_alignment')
parse_clipped_alignment.invoke('all', self._state_update % 'Parse clipped alignment')
# Input files
input_file = File(input_file)
# Output files
info = File('%s.info' % input_file.name)
self._info_files.append(info.name)
# Arguments
parse_clipped_alignment.addArguments(input_file)
parse_clipped_alignment.setStdout(info)
# Uses
parse_clipped_alignment.uses(input_file, link=Link.INPUT)
parse_clipped_alignment.uses(info, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(parse_clipped_alignment)
def _merge_stats(self):
merge_stats = Job(name='merge-stats')
merge_stats.invoke('all', self._state_update % 'Merging adaptor stats file')
# Outputs
adaptor_stats = File('%s.adaptor.stats' % self._prefix)
# Arguments
merge_stats.addArguments('reads*.stats', adaptor_stats)
for i in range(self._splits):
# Inputs
stats_i = File('reads%d.stats' % i)
# Uses
merge_stats.uses(stats_i, link=Link.INPUT)
# Outputs
merge_stats.uses(adaptor_stats, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(merge_stats)
def _parse_alignment(self, input_file, tag):
parse_alignment = Job(name='parse_alignment')
parse_alignment.invoke('all', self._state_update % 'Parse alignment')
# Input files
input_file = File(input_file)
# Output files
vis = File('%s.vis' % input_file.name)
self._vis_files.append(vis.name)
# Arguments
parse_alignment.addArguments(input_file, '--strandRule', self._strand_rule, '--tag', tag)
parse_alignment.setStdout(vis)
# Uses
parse_alignment.uses(input_file, link=Link.INPUT)
parse_alignment.uses(vis, link=Link.OUTPUT, transfer=False, register=False)
self.adag.addJob(parse_alignment)
def _merge_info(self, info_files, gtf_file):
merge_info = Job(name='merge-info')
merge_info.invoke('all', self._state_update % 'Merging info files to generate GTF file')
# Outputs
gtf_file = File(gtf_file)
for info_file in info_files:
# Inputs
info_i = File(info_file)
# Arguments
merge_info.addArguments(info_i)
# Uses
merge_info.uses(info_i, link=Link.INPUT)
# Arguments
merge_info.addArguments(gtf_file)
# Outputs
merge_info.uses(gtf_file, link=Link.OUTPUT, transfer=True, register=False)
self.adag.addJob(merge_info)
