Merge pull request #107 from yhoogstrate/additional_filter_intercept

Adds new type of filter that reduces FP ratio
yhoogstrate · Jan 25, 2018 · 912d77a · 912d77a
2 parents 1a201c3 + be13f87
commit 912d77a
Show file tree

Hide file tree

Showing 12 changed files with 74 additions and 58 deletions.
diff --git a/Changelog b/Changelog
@@ -1,3 +1,6 @@
+2018-01-25	Youri Hoogstrate	v0.15.1
+	 * Added filter lr/intercept
+
 2018-01-09	Youri Hoogstrate	v0.15.0
 	 * Bugfix resulting in higher number of detected frame shifts
 	 * `dr-disco integrate --fasta <fa file>` provides edit distance to

diff --git a/drdisco/DetectOutput.py b/drdisco/DetectOutput.py
@@ -428,6 +428,13 @@ def classify_intronic_exonic():
                     if chim_overhang < min_chim_overhang:
                         status.append("chim_overhang=" + str(chim_overhang) + "<" + str(min_chim_overhang))
 
+                    # @todo subfunc
+                    lr_intercept_max = (-31.0 * ((e.score + 100.0) / (450.0 + e.score + 100.0))) + 85.25
+                    if e.lr_A_intercept > lr_intercept_max:
+                        status.append("lr_A_intercept=" + str(e.lr_A_intercept) + ">" + str(lr_intercept_max))
+                    if e.lr_B_intercept > lr_intercept_max:
+                        status.append("lr_B_intercept=" + str(e.lr_B_intercept) + ">" + str(lr_intercept_max))
+
                     if len(status) == 0:
                         e.status = 'valid'
                         fh.write(str(e))

diff --git a/drdisco/__init__.py b/drdisco/__init__.py
@@ -31,7 +31,7 @@
 import logging
 import sys
 
-__version_info__ = ('0', '15', '0')
+__version_info__ = ('0', '15', '1')
 __version__ = '.'.join(__version_info__) if (len(__version_info__) == 3) else '.'.join(__version_info__[0:3]) + "-" + __version_info__[3]
 __author__ = 'Youri Hoogstrate'
 __homepage__ = 'https://github.com/yhoogstrate/dr-disco'

diff --git a/share/blacklist-regions.hg38.bed b/share/blacklist-regions.hg38.bed
@@ -3476,3 +3476,9 @@ chr17	82079668	82079707	+	poly-T
 chr17	82079668	82079707	-	poly-T
 chr5	141989680	141989720	+	poly-A
 chr5	141989680	141989720	-	poly-A
+chr16	78758908	78758976	+	poly-A
+chr16	78758908	78758976	-	poly-A
+
+chr8:102,365,441-102,365,508
+chr3:123,143,378-123,143,417
+
diff --git a/tests/classify/test_11.out.dbed b/tests/classify/test_11.out.dbed
@@ -1,2 +1,2 @@
 chr-A	pos-A	direction-A	pos-A-acceptor	pos-A-donor	chr-B	pos-B	direction-B	pos-B-acceptor	pos-B-donor	genomic-distance	filter-status	circRNA	intronic/exonic	score	soft+hardclips	n-split-reads	n-discordant-reads	alignment-score	mismatches	n-edges	n-nodes-A	n-nodes-B	n-splice-junc-A	n-splice-junc-B	entropy-bp-edge	entropy-all-edges	bp-pos-stddev	entropy-disco-bps	lr-A-slope	lr-A-intercept	lr-A-rvalue	lr-A-pvalue	lr-A-stderr	lr-B-slope	lr-B-intercept	lr-B-rvalue	lr-B-pvalue	lr-B-stderr	disco/split	clips/score	nodes/edge	median-AS-A	median-AS-B	max-AS-A	max-AS-B	data-structure
-chr7	151970252	+	0	2	chr21	39777764	-	2	0	inf	entropy=0.0<0.8394,n_support=1<7,n_lr_symmetry=112.87>=24.68	linear	intronic	3	2	1	0	246	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	14.0000	0.0000	1.0000	0.0000	0.0000	112.0000	0.0000	1.0000	0.0000	0.0000	0.3333	2.0000	14	112	14	112	chr7:151970252/151970253(+)->chr21:39777764/39777765(-):(spanning_paired_1_t:1,spanning_paired_2_t:1)
+chr7	151970252	+	0	2	chr21	39777764	-	2	0	inf	entropy=0.0<0.8394,n_support=1<7,n_lr_symmetry=112.87>=24.68,lr_B_intercept=112.0>79.476039783	linear	intronic	3	2	1	0	246	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	14.0000	0.0000	1.0000	0.0000	0.0000	112.0000	0.0000	1.0000	0.0000	0.0000	0.3333	2.0000	14	112	14	112	chr7:151970252/151970253(+)->chr21:39777764/39777765(-):(spanning_paired_1_t:1,spanning_paired_2_t:1)
diff --git a/tests/classify/test_12.out.dbed b/tests/classify/test_12.out.dbed
@@ -1,3 +1,3 @@
 chr-A	pos-A	direction-A	pos-A-acceptor	pos-A-donor	chr-B	pos-B	direction-B	pos-B-acceptor	pos-B-donor	genomic-distance	filter-status	circRNA	intronic/exonic	score	soft+hardclips	n-split-reads	n-discordant-reads	alignment-score	mismatches	n-edges	n-nodes-A	n-nodes-B	n-splice-junc-A	n-splice-junc-B	entropy-bp-edge	entropy-all-edges	bp-pos-stddev	entropy-disco-bps	lr-A-slope	lr-A-intercept	lr-A-rvalue	lr-A-pvalue	lr-A-stderr	lr-B-slope	lr-B-intercept	lr-B-rvalue	lr-B-pvalue	lr-B-stderr	disco/split	clips/score	nodes/edge	median-AS-A	median-AS-B	max-AS-A	max-AS-B	data-structure
-chr21	39817544	-	4	0	chr21	42880007	+	0	4	3062463	entropy=0.0<0.8341,n_support=2<7,n_lr_symmetry=103.08>=17.71	linear	intronic	6	4	2	0	496	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	100.0000	0.0000	1.0000	0.0000	0.0000	25.0000	0.0000	1.0000	0.0000	0.0000	0.3333	2.0000	100	25	100	25	chr21:39817544/39817545(-)->chr21:42880007/42880008(+):(spanning_paired_1:2,spanning_paired_2:2)
+chr21	39817544	-	4	0	chr21	42880007	+	0	4	3062463	entropy=0.0<0.8341,n_support=2<7,n_lr_symmetry=103.08>=17.71,lr_A_intercept=100.0>79.3399280576	linear	intronic	6	4	2	0	496	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	100.0000	0.0000	1.0000	0.0000	0.0000	25.0000	0.0000	1.0000	0.0000	0.0000	0.3333	2.0000	100	25	100	25	chr21:39817544/39817545(-)->chr21:42880007/42880008(+):(spanning_paired_1:2,spanning_paired_2:2)
 chr21	39792184	-	6	0	chr21	42879912	+	0	6	3087728	n_support=3<8	linear	intronic	3	0	0	3	726	0	3	3	1	2	0	0.0000	1.0000	0.0000	0.0000	0.0000	4.0000	0.0000	1.0000	0.0000	0.0000	4.0000	0.0000	1.0000	0.0000	60.0000	0.0000	1.3333	126	120	126	120	chr21:39792184/39792185(-)->chr21:42879912/42879913(+):(discordant_mates:2)&chr21:39793115/39793116(-)->chr21:42879912/42879913(+):(discordant_mates:2)&chr21:39795473/39795474(-)->chr21:42879912/42879913(+):(discordant_mates:2)
diff --git a/tests/classify/test_13.out.dbed b/tests/classify/test_13.out.dbed
@@ -5,5 +5,5 @@ chr21	40073871	-	18	0	chr21	42870045	+	0	18	2796174	valid	linear	intronic	26	103
 chr21	40056045	-	14	0	chr21	42870045	+	0	14	2814000	valid	linear	intronic	20	92	6	2	1850	2	3	2	3	0	1	1.0000	1.0000	0.0000	0.4056	13.4857	46.9524	0.8602	0.0280	3.9977	12.8000	12.3333	0.8708	0.0240	3.6137	0.3333	2.3000	1.6667	94	31	126	126	chr21:40056045/40056046(-)->chr21:42870045/42870046(+):(discordant_mates:2,spanning_paired_1:5,spanning_paired_2:5)&chr21:40056045/40056046(-)->chr21:42880007/42880008(+):(spanning_paired_1:1,spanning_paired_2:1)&chr21:40055887/40055888(-)->chr21:42878380/42878381(+):(discordant_mates:2)
 chr21	39817544	-	8	0	chr21	42878371	+	0	8	3060827	n_support=10<12,log_ratio_slope=13.2>1.8,log_ratio_rvalue=9.21>0.4	linear	exonic	18	33	4	6	1886	1	5	5	4	1	0	1.0000	0.9398	0.0000	0.6395	54.0000	14.0000	1.0000	0.0000	0.0000	0.0000	41.0000	0.0000	1.0000	0.0000	1.5000	0.9167	1.8000	73	51	126	116	chr21:39817544/39817545(-)->chr21:42878371/42878372(+):(spanning_paired_1:2,spanning_paired_2:2)&chr21:40064721/40064722(-)->chr21:42878371/42878372(+):(spanning_paired_1:2,spanning_paired_2:2)&chr21:39817370/39817371(-)->chr21:42883791/42883792(+):(discordant_mates:8)&chr21:39817410/39817411(-)->chr21:42878907/42878908(+):(discordant_mates:2)&chr21:39817510/39817511(-)->chr21:42883787/42883788(+):(discordant_mates:2)
 chr21	39850904	-	8	0	chr21	42877894	+	0	8	3026990	n_support=4<7	linear	intronic	8	4	2	2	895	0	1	1	1	0	0	1.0000	1.0000	0.0000	1.0000	37.2000	23.7000	0.9039	0.0961	12.4491	16.3000	70.3000	0.9316	0.0684	4.4978	1.0000	0.2500	2.0000	35	90	126	126	chr21:39850904/39850905(-)->chr21:42877894/42877895(+):(discordant_mates:4,spanning_paired_1_t:2,spanning_paired_2_t:2)
-chr21	39969017	-	2	0	chr21	42870045	+	0	2	2901028	n_support=2<8,n_lr_symmetry=105.32>=24.68	linear	intronic	3	75	1	1	333	0	2	2	2	0	0	0.0000	1.0000	0.0000	0.0000	0.0000	103.0000	0.0000	1.0000	0.0000	0.0000	22.0000	0.0000	1.0000	0.0000	1.0000	12.5000	2.0000	103	22	124	91	chr21:39969017/39969018(-)->chr21:42870045/42870046(+):(spanning_singleton_1:1,spanning_singleton_2:1)&chr21:39968827/39968828(-)->chr21:42879876/42879877(+):(discordant_mates:2)
+chr21	39969017	-	2	0	chr21	42870045	+	0	2	2901028	n_support=2<8,n_lr_symmetry=105.32>=24.68,lr_A_intercept=103.0>79.476039783	linear	intronic	3	75	1	1	333	0	2	2	2	0	0	0.0000	1.0000	0.0000	0.0000	0.0000	103.0000	0.0000	1.0000	0.0000	0.0000	22.0000	0.0000	1.0000	0.0000	1.0000	12.5000	2.0000	103	22	124	91	chr21:39969017/39969018(-)->chr21:42870045/42870046(+):(spanning_singleton_1:1,spanning_singleton_2:1)&chr21:39968827/39968828(-)->chr21:42879876/42879877(+):(discordant_mates:2)
 chr21	40103950	+	0	2	chr21	42869215	-	2	0	2765265	entropy=0.0<0.8394,n_support=1<7,n_lr_symmetry=73.25>=24.68	linear	intronic	3	2	1	0	162	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	46.0000	0.0000	1.0000	0.0000	0.0000	57.0000	0.0000	1.0000	0.0000	0.0000	0.3333	2.0000	46	57	46	57	chr21:40103950/40103951(+)->chr21:42869215/42869216(-):(spanning_paired_1:1,spanning_paired_2:1)
diff --git a/tests/classify/test_15.out.dbed b/tests/classify/test_15.out.dbed
@@ -1,3 +1,3 @@
 chr-A	pos-A	direction-A	pos-A-acceptor	pos-A-donor	chr-B	pos-B	direction-B	pos-B-acceptor	pos-B-donor	genomic-distance	filter-status	circRNA	intronic/exonic	score	soft+hardclips	n-split-reads	n-discordant-reads	alignment-score	mismatches	n-edges	n-nodes-A	n-nodes-B	n-splice-junc-A	n-splice-junc-B	entropy-bp-edge	entropy-all-edges	bp-pos-stddev	entropy-disco-bps	lr-A-slope	lr-A-intercept	lr-A-rvalue	lr-A-pvalue	lr-A-stderr	lr-B-slope	lr-B-intercept	lr-B-rvalue	lr-B-pvalue	lr-B-stderr	disco/split	clips/score	nodes/edge	median-AS-A	median-AS-B	max-AS-A	max-AS-B	data-structure
-chr21	39817544	-	4	0	chr21	42870045	+	0	4	3052501	n_support=2<7,log_ratio_slope=2.4>1.8	linear	exonic	4	2	1	1	480	0	1	1	1	0	0	1.0000	1.0000	0.0000	1.0000	99.0000	24.0000	1.0000	0.0000	0.0000	9.0000	102.0000	1.0000	0.0000	0.0000	1.0000	0.2500	2.0000	24	102	123	111	chr21:39817544/39817545(-)->chr21:42870045/42870046(+):(discordant_mates:2,spanning_paired_1:1,spanning_paired_2:1)
-chr21	39956806	-	4	0	chr21	42870051	+	0	4	2913245	entropy=0.0<0.841,n_support=2<7,n_lr_symmetry=176.07>=28.44	linear	intronic	2	0	0	2	492	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	124.0000	0.0000	1.0000	0.0000	0.0000	125.0000	0.0000	1.0000	0.0000	40.0000	0.0000	2.0000	124	125	124	125	chr21:39956806/39956807(-)->chr21:42870051/42870052(+):(discordant_mates:4)
+chr21	39817544	-	4	0	chr21	42870045	+	0	4	3052501	n_support=2<7,log_ratio_slope=2.4>1.8,lr_B_intercept=102.0>79.4305054152	linear	exonic	4	2	1	1	480	0	1	1	1	0	0	1.0000	1.0000	0.0000	1.0000	99.0000	24.0000	1.0000	0.0000	0.0000	9.0000	102.0000	1.0000	0.0000	0.0000	1.0000	0.2500	2.0000	24	102	123	111	chr21:39817544/39817545(-)->chr21:42870045/42870046(+):(discordant_mates:2,spanning_paired_1:1,spanning_paired_2:1)
+chr21	39956806	-	4	0	chr21	42870051	+	0	4	2913245	entropy=0.0<0.841,n_support=2<7,n_lr_symmetry=176.07>=28.44,lr_A_intercept=124.0>79.5217391304,lr_B_intercept=125.0>79.5217391304	linear	intronic	2	0	0	2	492	0	1	1	1	0	0	0.0000	0.0000	0.0000	0.0000	0.0000	124.0000	0.0000	1.0000	0.0000	0.0000	125.0000	0.0000	1.0000	0.0000	40.0000	0.0000	2.0000	124	125	124	125	chr21:39956806/39956807(-)->chr21:42870051/42870052(+):(discordant_mates:4)