resolved conflicts after merge

Author: akahles

.circleci/config.yml

@@ -2,7 +2,7 @@ version: 2
 jobs:
   build:
     docker:
-      - image: circleci/python:3.6.8-jessie
+      - image: cimg/python:3.9.8-node
     steps:
       - checkout
       - run:

Makefile

@@ -76,7 +76,7 @@ docs: ## generate Sphinx HTML documentation, including API docs
 	sphinx-apidoc -o docs/ spladder
 	$(MAKE) -C docs clean
 	$(MAKE) -C docs html
-	$(BROWSER) docs/_build/html/index.html
+	$(BROWSER) docs/build/html/index.html
 
 servedocs: docs ## compile the docs watching for changes
 	watchmedo shell-command -p '*.rst' -c '$(MAKE) -C docs html' -R -D .

docs/source/conf.py

@@ -24,9 +24,9 @@
 author = u'Andre Kahles'
 
 # The short X.Y version
-version = u'2.4'
+version = u'2.5'
 # The full version, including alpha/beta/rc tags
-release = u'2.4.2'
+release = u'2.5.0'
 
 
 # -- General configuration ---------------------------------------------------

docs/source/file_formats.rst

@@ -121,7 +121,7 @@ In its latest version, SplAdder also supports (on an experimental level) CRAM co
 files as input. If you are using such files, in addition to the input filenames of the
 alignment files, also the path to the indexed reference sequence used for compression is required::
 
-    spladder build --bams alignment1.cram,alignment2.cram,... --cram-reference path/to/cram_ref.fa 
+    spladder build --bams alignment1.cram,alignment2.cram,... --reference path/to/cram_ref.fa 
 
 **Alignment**
     By default, SplAdder only uses primary alignments (in SAM/BAM the ones not carrying the 256
@@ -319,27 +319,107 @@ entirely (for instance to carry it out at a later point in time). This is done v
 
     spladder build ... --no-extract-ase ...
 
-SplAdder can currently extract 6 different types of alternative splicing events:
+**Event extraction**
+    SplAdder can currently extract 6 different types of alternative splicing events:
 
-- exon skips (`exon_skip`)
-- intron retentions (`intron_retention`)
-- alternative 3' splice sites (`alt_3prime`)
-- alternative 5' splice sites (`alt_5prime`)
-- mutually exclusive exons (`mutex_exons`)
-- multiple (coordinated) exons skips (`mult_exon_skip`)
+    - exon skips (`exon_skip`)
+    - intron retentions (`intron_retention`)
+    - alternative 3' splice sites (`alt_3prime`)
+    - alternative 5' splice sites (`alt_5prime`)
+    - mutually exclusive exons (`mutex_exons`)
+    - multiple (coordinated) exons skips (`mult_exon_skip`)
 
-Per default all events of all types are extracted from the graph. To specify a single type or a
-subset of types (e.g., exon skips and mutually exclusive exons only), the user can specify the short
-names of the event types (as shown in parentheses above) as follows::
+    Per default all events of all types are extracted from the graph. To specify a single type or a
+    subset of types (e.g., exon skips and mutually exclusive exons only), the user can specify the short
+    names of the event types (as shown in parentheses above) as follows::
 
-    spladder build ... --event-types exon_skip,mutex_exons ...
+        spladder build ... --event-types exon_skip,mutex_exons ...
+
+    In some cases (for instance when integrating hundreds of alignment samples), the splicing graphs can
+    grow very complex. To limit the running time, an upper bound for the maximum number of edges in the
+    splicing graph of a gene to be used for event extraction is set. This threshold is 500 per default.
+    To adapt this threshold, e.g., to 250, the user can specify::
+        
+        spladder build ... --ase-edge-limit 250 ...
+
+**Event verification**
+    Similar to graph validation, SplAdder also performs a step of splice event verification. Only
+    verified events are reported as confident to the user. There are two possibilities how the validity of
+    a confident event is established.
+
+    The classical way for event verification is to use heuristic criteria based on the RNA-Seq
+    evidence provided to SplAdder. Depending on the alternative event type, as different set of
+    criteria is used. The tables below summarize the criteria currently in use for the different
+    event types. The order and numbering of criteria is the same as used in the output files of
+    SplAdder.
+
+    +----------------------------------------------------------------------------------------+
+    | Multiple Exon Skip                                                                     |
+    +====+===================================================================================+
+    | 0  | exon coordinates are valid (>= 0 && start < stop && non-overlapping) &            | 
+    |    | skipped exon coverage >= FACTOR * mean(pre, after)                                |
+    +----+-----------------------------------------------------------------------------------+
+    | 1  | inclusion count first intron >= threshold                                         |
+    +----+-----------------------------------------------------------------------------------+
+    | 2  | inclusion count last intron >= threshold                                          |
+    +----+-----------------------------------------------------------------------------------+
+    | 3  | avg inclusion count inner exons >= threshold                                      |
+    +----+-----------------------------------------------------------------------------------+
+    | 4  | skip count >= threshold                                                           |
+    +----+-----------------------------------------------------------------------------------+ 
+
+    +----+-----------------------------------------------------------------------------------+ 
+    | Intron Retention                                                                       |
+    +====+===================================================================================+
+    | 0  | counts meet criteria for min_retention_cov, min_retention_region and              |
+    |    | min_retetion_rel_cov                                                              |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 1  | min_non_retention_count >= threshold                                              |
+    +----+-----------------------------------------------------------------------------------+ 
+
+    +----+-----------------------------------------------------------------------------------+ 
+    | Exon Skip                                                                              |
+    +====+===================================================================================+
+    | 0  | coverage of skipped exon is >= than FACTOR * mean(pre, after)                     |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 1  | inclusion count of first intron >= threshold                                      |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 2  | inclusion count of second intron >= threshold                                     |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 3  | skip count of exon >= threshold                                                   |
+    +----+-----------------------------------------------------------------------------------+ 
+
+    +----+-----------------------------------------------------------------------------------+ 
+    | Alternative 3/5 Prime                                                                  |
+    +====+===================================================================================+
+    | 0  | coverage of diff region is at least FACTOR * coverage constant region             |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 1  | both alternative introns are >= threshold                                         |
+    +----+-----------------------------------------------------------------------------------+ 
+
+    +----+-----------------------------------------------------------------------------------+ 
+    | Mutually Exclusive Exons                                                               |
+    +====+===================================================================================+
+    | 0  | coverage of first alt exon is >= than FACTOR times average of pre and after       |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 1  | coverage of second alt exon is >= than FACTOR times average of pre and after      |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 2  | both introns neighboring first alt exon are confirmed >= threshold                |
+    +----+-----------------------------------------------------------------------------------+ 
+    | 3  | both introns neighboring second alt exon are confirmed >= threshold               |
+    +----+-----------------------------------------------------------------------------------+ 
+
+    In addition to the classical, RNA-Seq evidence based mode, since version 2.5 it is also allowed
+    to use the provided annotation to verify an existing event. In this mode each one of the
+    criteria listed above is replaced with a lookup in the provided annotation. That is, if an
+    intron is already annotated, it will be used for event verification irrespective of any RNA-Seq
+    expression support. This mode is especially useful for single sample analysis, where a complete
+    isoform switch might have occurred and only the alternative event path is supported by reads but
+    not the annotated one. In this case, the event is still reported. This mode is switched off by
+    default and can be activated via::
+
+        spladder build ... -use-anno-support ...
 
-In some cases (for instance when integrating hundreds of alignment samples), the splicing graphs can
-grow very complex. To limit the running time, an upper bound for the maximum number of edges in the
-splicing graph of a gene to be used for event extraction is set. This threshold is 500 per default.
-To adapt this threshold, e.g., to 250, the user can specify::
-    
-    spladder build ... --ase-edge-limit 250 ...
 
 The ``test`` mode
 -----------------

docs/source/img/splice_events.pdf

@@ -1,4 +1,5 @@
 numpy>=1.14.6
+numba>=0.52.0
 matplotlib>=2.2
 scipy>=1.3
 intervaltree>=3.0.0

docs/source/img/splice_events.png

@@ -15,17 +15,19 @@ for i in $(seq 1 20)
 do
     bams="$bams,${datadir}/align/testcase_${testname}_1_sample${i}.bam"
 done
-python -m spladder.spladder build -v -o ${outdir} -a ${datadir}/testcase_${testname}_spladder.gtf -b ${bams#,} --event-types exon_skip,intron_retention,alt_3prime,alt_5prime,mutex_exons,mult_exon_skip --readlen 50 --output-conf-icgc --output-txt --output-txt-conf --output-gff3 --output-struc --output-struc-conf --output-bed --output-conf-bed --output-conf-tcga
+python -m spladder.spladder build -v -o ${outdir} -a ${datadir}/testcase_${testname}_spladder.gtf -b ${bams#,} --event-types exon_skip,intron_retention,alt_3prime,alt_5prime,mutex_exons,mult_exon_skip --readlen 50 --output-conf-icgc --output-txt --output-txt-conf --output-gff3 --output-struc --output-struc-conf --output-bed --output-conf-bed --output-conf-tcga #--use-anno-support
 
 bamsA=bamlistA.txt
+rm -f $bamsA
 for i in $(seq 1 10)
 do
     echo "${datadir}/align/testcase_${testname}_1_sample${i}.bam" >> $bamsA
 done
 bamsB=bamlistB.txt
+rm -f $bamsB
 for i in $(seq 11 20)
 do
     echo "align/testcase_${testname}_1_sample${i}.bam" >> $bamsB
 done
-python -m spladder.spladder test -o ${outdir} -v --diagnose-plots -f ps --readlen 50 --merge-strat merge_graphs --event-types exon_skip -a $bamsA -b $bamsB
+python -m spladder.spladder test -o ${outdir} -v --diagnose-plots -f pdf --readlen 50 --merge-strat merge_graphs --event-types exon_skip -a $bamsA -b $bamsB --dpsi 0
 rm bamlistA.txt bamlistB.txt

docs/source/spladder_modes.rst

@@ -19,17 +19,19 @@ do
     bams="$bams,${datadir}/align/testcase_${testname}_1_sample${i}.bam"
 done
 #export REF_PATH=$genome
-python -m spladder.spladder build -v -o ${outdir} -a ${datadir}/testcase_${testname}_spladder.gtf -b ${crams#,} --event-types exon_skip,intron_retention,alt_3prime,alt_5prime,mutex_exons,mult_exon_skip --readlen 50 --output-conf-icgc --output-txt --output-txt-conf --output-gff3 --output-struc --output-struc-conf --output-bed --output-conf-bed --output-conf-tcga --cram-ref ${genome}
+python -m spladder.spladder build -v -o ${outdir} -a ${datadir}/testcase_${testname}_spladder.gtf -b ${crams#,} --event-types exon_skip,intron_retention,alt_3prime,alt_5prime,mutex_exons,mult_exon_skip --readlen 50 --output-conf-icgc --output-txt --output-txt-conf --output-gff3 --output-struc --output-struc-conf --output-bed --output-conf-bed --output-conf-tcga --reference ${genome}
 
 cramsA=cramlistA.txt
-for i in $(seq 1 10)
+rm -f $cramsA
+for i in 10 8 2 1 7 6 5 3 9 4
 do
     echo "${datadir}/align/testcase_${testname}_1_sample${i}.cram" >> $cramsA
 done
 cramsB=cramlistB.txt
-for i in $(seq 11 20)
+rm -f $cramsB
+for i in 20 13 17 11 12 19 15 14 16 18
 do
     echo "align/testcase_${testname}_1_sample${i}.cram" >> $cramsB
 done
-python -m spladder.spladder test -o ${outdir} -v --diagnose-plots -f ps --readlen 50 --merge-strat merge_graphs --event-types exon_skip -a $cramsA -b $cramsB
+python -m spladder.spladder test -o ${outdir} -v --diagnose-plots -f pdf --readlen 50 --merge-strat merge_graphs --event-types exon_skip -a $cramsA -b $cramsB --dpsi 0
 rm cramlistA.txt cramlistB.txt