bugfix for large D in cigars

vpc-ccg · Apr 7, 2021 · b8f8dbb · b8f8dbb
1 parent 0285eb8
commit b8f8dbb
Show file tree

Hide file tree

Showing 7 changed files with 10 additions and 107 deletions.
diff --git a/README.md b/README.md
@@ -99,13 +99,13 @@ Align takes the following arguments:
 ### Isoforms
 
 ```
-py/freddie_isoforms.py --segment-dir {SEGMENT} --cluster-dir {CLUSTER} --output {ISOFORMS.GTF} -t {threads}
+py/freddie_isoforms.py --split-dir {SPLIT} --cluster-dir {CLUSTER} --output {ISOFORMS.GTF} -t {threads}
 ```
 
 Align takes the following arguments:
 
-- `--segment-dir/-s`: `SEGMENT` output directory of the segment stage
-- `--cluster-dir/-s`: `CLUSTER` output directory of the cluster stage
+- `--split-dir/-s`: `SPLIT` output directory of the segment stage
+- `--cluster-dir/-c`: `CLUSTER` output directory of the cluster stage
 - `--output/-o`: Output GTF file of isoforms stage. Default: `freddie_isoforms.gtf`
 - `--threads/-t`: Number of threads. Default: 1
 

diff --git a/Snakefile b/Snakefile
@@ -100,7 +100,7 @@ rule cluster:
 rule isoforms:
     input:
         script  = config['exec']['isoforms'],
-        segment = '{}/{{sample}}/freddie.segment'.format(output_d),
+        split   = '{}/{{sample}}/freddie.split'.format(output_d),
         cluster = '{}/{{sample}}/freddie.cluster'.format(output_d),
     output:
         isoforms = protected('{}/{{sample}}/freddie.isoforms.gtf'.format(output_d)),
@@ -112,4 +112,4 @@ rule isoforms:
         mem  = "16G",
         time = 359,
     shell:
-        '{input.script} -s {input.segment} -c {input.cluster} -o {output.isoforms} -t {threads}'
+        '{input.script} -s {input.split} -c {input.cluster} -o {output.isoforms} -t {threads}'
diff --git a/config.yaml b/config.yaml
@@ -3,7 +3,7 @@ outpath:
 
 gurobi:
     license: gurobi.lic
-    timeout: 15
+    timeout: 1
 
 exec:
     split    : py/freddie_split.py

diff --git a/py/freddie_align.py b/py/freddie_align.py
diff --git a/py/freddie_isoforms.py b/py/freddie_isoforms.py
diff --git a/py/freddie_segment.py b/py/freddie_segment.py
@@ -313,8 +313,8 @@ def get_unaligned_gaps_and_polyA(read, segs):
     (_, l_seg_idx) = intervals[-1]
     end = segs[l_seg_idx][1]
     q_esc_pos, _ = get_interval_end(end=end, read=read)
-    assert 0 <= q_ssc_pos < q_esc_pos <= read['length'], (
-        q_ssc_pos, q_esc_pos, read['length'], start, end, read)
+    assert 0 <= q_ssc_pos <= q_esc_pos <= read['length'], (
+        q_ssc_pos, q_esc_pos, read['length'], start, end, segs, read)
     s_polys = list()
     for char in ['A', 'T']:
         s = 0

diff --git a/py/freddie_split.py b/py/freddie_split.py
@@ -142,7 +142,6 @@ def get_intervals(aln):
 
     for t, c in cigar:
         assert(0 <= t < 10)
-    for t, c in cigar:
         if t in target_consuming:
             tend += c
     qstart_c = qstart
@@ -153,6 +152,8 @@ def get_intervals(aln):
     intervals = list()
     interval_cigar = list()
     for t, c in cigar:
+        if t == pysam.CDEL and c > 20:
+            t = pysam.CREF_SKIP
         if t in exon_consuming:
             interval_cigar.append((t, c))
         if t == pysam.CDEL: