Mercurial > repos > davidvanzessen > shm_csr

changeset 98:d714f5ea83d7 draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
planemo upload commit 1a01065a084a817382872154f779b94090a35ebf
author rhpvorderman
date Wed, 10 Jan 2024 12:32:47 +0000
parents fbc6307dd83b
children
files CHANGELOG.md CONTROL_NWK377_PB_IGHC_MID1_40nt_2.txz a.out baseline/wrapper.sh shm_csr.xml show_time_as_float tests/.pytest_cache/.gitignore tests/.pytest_cache/CACHEDIR.TAG tests/.pytest_cache/README.md tests/.pytest_cache/v/cache/lastfailed tests/.pytest_cache/v/cache/nodeids tests/.pytest_cache/v/cache/stepwise tests/__pycache__/test_shm_csr.cpython-37-pytest-6.2.5.pyc tests/__pycache__/test_shm_csr.cpython-39-pytest-7.2.1.pyc tests/__pycache__/test_shm_csr.cpython-39-pytest-7.4.4.pyc tests/data/.~lock.handleiding activeren pas.docx# tests/data/handleiding activeren pas.docx tests/test_shm_csr.py time_ns wrapper.sh
diffstat 20 files changed, 105 insertions(+), 57 deletions(-) [+]
line wrap: on
line diff
--- a/CHANGELOG.md	Mon Jan 08 08:50:07 2024 +0000
+++ b/CHANGELOG.md	Wed Jan 10 12:32:47 2024 +0000
@@ -1,3 +1,8 @@
+version 1.10.0
+------------------
++ Fix crashing baseline when only unproductive sequences are present.
++ Replace "fast" option with options to toggle baseline and Change-O indidually.
+
 version 1.9.1
 -----------------
 + Fix error in launch command.
Binary file CONTROL_NWK377_PB_IGHC_MID1_40nt_2.txz has changed
Binary file a.out has changed
--- a/baseline/wrapper.sh	Mon Jan 08 08:50:07 2024 +0000
+++ b/baseline/wrapper.sh	Wed Jan 10 12:32:47 2024 +0000
@@ -1,4 +1,5 @@
 #!/bin/bash
+
 dir="$(cd "$(dirname "$0")" && pwd)"
 
 testID=$1
@@ -79,14 +80,16 @@
 
 echo "$workdir/${outID}.txt"
 
-rows=`tail -n +2 $workdir/${outID}.txt | grep -v "All sequences combined" | grep -n 'Group' | grep -Eoh '^[0-9]+' | tr '\n' ' '`
-rows=($rows)
-#unset rows[${#rows[@]}-1]
+if [[ -f "$workdir/${outID}.txt" ]]
+then
+  rows=`tail -n +2 $workdir/${outID}.txt | grep -v "All sequences combined" | grep -n 'Group' | grep -Eoh '^[0-9]+' | tr '\n' ' '`
+  rows=($rows)
+  #unset rows[${#rows[@]}-1]
 
-cd $dir
-Rscript --verbose $dir/comparePDFs.r $workdir/${outID}.RData $output ${rows[@]} 2>&1
-cp $workdir/result.txt ${output_table}
+  cd $dir
+  Rscript --verbose $dir/comparePDFs.r $workdir/${outID}.RData $output ${rows[@]} 2>&1
+  cp $workdir/result.txt ${output_table}
+fi
 
 
 
-
--- a/shm_csr.xml	Mon Jan 08 08:50:07 2024 +0000
+++ b/shm_csr.xml	Wed Jan 10 12:32:47 2024 +0000
@@ -1,4 +1,4 @@
-<tool id="shm_csr" name="SHM &amp; CSR pipeline" version="1.9.1" profile="16.04">
+<tool id="shm_csr" name="SHM &amp; CSR pipeline" version="1.10.0" profile="16.04">
 	<description></description>
 	<requirements>
 		<requirement type="package" version="3.7.1">python</requirement>
@@ -35,7 +35,7 @@
 			$filter_unique.filter_unique_clone_count
 			$class_filter_cond.class_filter
 			$empty_region_filter
-			$fast
+			$changeo $baseline
 		#else:
 			$__tool_directory__/wrapper.sh "$in_file"
 			custom
@@ -48,7 +48,7 @@
 			2
 			$class_filter_cond.class_filter
 			$empty_region_filter
-			$fast
+			$changeo $baseline
 		#end if
 	]]>
 	</command>
@@ -108,59 +108,57 @@
 			</param>
 		</conditional>
 		<conditional name="naive_output_cond">
-			<param name="naive_output" type="select" label="Output new IMGT archives per class into your history?">
-				<option value="yes">Yes</option>
-				<option value="no" selected="true">No</option>
-			</param>
-			<when value="yes"></when>
-			<when value="no"></when>
-		</conditional>
-		<param name="fast" type="select" label="Fast" help="Skips Change-O/Baseline" >
-			<option value="yes">Yes</option>
-			<option value="no" selected="true">No</option>
-		</param>
+			<param name="naive_output" type="boolean" label="Output new IMGT archives per class into your history?"
+				   checked="no" truevalue="yes" falsevalue="no"/>
+			</conditional>
+		<param name="changeo" type="boolean" label="Run Change-O"
+			   checked="yes" truevalue="yes" falsevalue="no"/>
+		<param name="baseline" type="boolean" label="Run Baseline"
+			   checked="yes" truevalue="yes" falsevalue="no"
+			   help="Baseline is automatically skipped when functionality filter is set to unproductive only."
+		/>
 	</inputs>
 	<outputs>
 		<data format="html" name="out_file" label = "SHM &amp; CSR on ${in_file.name}"/>
 		<data format="imgt_archive" name="naive_output_ca" label = "Filtered IMGT IGA: ${in_file.name}" >
-		    <filter>naive_output_cond['naive_output'] == "yes"</filter>
+		    <filter>naive_output_cond['naive_output'] is True</filter>
 		    <filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGG"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGM"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_cg" label = "Filtered IMGT IGG: ${in_file.name}" >
-		    <filter>naive_output_cond['naive_output'] == "yes"</filter>
+		    <filter>naive_output_cond['naive_output'] is True</filter>
 		    <filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGA"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGM"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_cm" label = "Filtered IMGT IGM: ${in_file.name}" >
-		    <filter>naive_output_cond['naive_output'] == "yes"</filter>
+		    <filter>naive_output_cond['naive_output'] is True</filter>
 		    <filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGA"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGG"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_ce" label = "Filtered IMGT IGE: ${in_file.name}" >
-		    <filter>naive_output_cond['naive_output'] == "yes"</filter>
+		    <filter>naive_output_cond['naive_output'] is True</filter>
 		    <filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGA"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGG"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGM"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_igm_naive" label = "Filtered naive IGM sequences (mutations below 2%): ${in_file.name}">
-			<filter>naive_output_cond['naive_output'] == "yes"</filter>
+			<filter>naive_output_cond['naive_output'] is True</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGA"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGG"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_igm_naive_memory" label = "Filtered naive memory IGM sequences (mutations over 2%): ${in_file.name}">
-			<filter>naive_output_cond['naive_output'] == "yes"</filter>
+			<filter>naive_output_cond['naive_output'] is True</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGA"</filter>
 			<filter>class_filter_cond['class_filter'] != "101_101_IGG"</filter>
 		</data>
 		<data format="imgt_archive" name="naive_output_all" label = "Filtered IMGT all: ${in_file.name}" >
-		    <filter>naive_output_cond['naive_output'] == "yes"</filter>
+		    <filter>naive_output_cond['naive_output'] is True</filter>
 		    <filter>class_filter_cond['class_filter'] == "101_101"</filter>
 		</data>
 	</outputs>
Binary file show_time_as_float has changed
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/.gitignore	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,2 @@
+# Created by pytest automatically.
+*
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/CACHEDIR.TAG	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,4 @@
+Signature: 8a477f597d28d172789f06886806bc55
+# This file is a cache directory tag created by pytest.
+# For information about cache directory tags, see:
+#	https://bford.info/cachedir/spec.html
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/README.md	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,8 @@
+# pytest cache directory #
+
+This directory contains data from the pytest's cache plugin,
+which provides the `--lf` and `--ff` options, as well as the `cache` fixture.
+
+**Do not** commit this to version control.
+
+See [the docs](https://docs.pytest.org/en/stable/how-to/cache.html) for more information.
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/v/cache/lastfailed	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,3 @@
+{
+  "test_shm_csr.py::test_baseline_succeeds": true
+}
\ No newline at end of file
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/v/cache/nodeids	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,3 @@
+[
+  "test_shm_csr.py::test_baseline_succeeds"
+]
\ No newline at end of file
--- /dev/null	Thu Jan 01 00:00:00 1970 +0000
+++ b/tests/.pytest_cache/v/cache/stepwise	Wed Jan 10 12:32:47 2024 +0000
@@ -0,0 +1,1 @@
+[]
\ No newline at end of file
Binary file tests/__pycache__/test_shm_csr.cpython-37-pytest-6.2.5.pyc has changed
Binary file tests/__pycache__/test_shm_csr.cpython-39-pytest-7.2.1.pyc has changed
Binary file tests/__pycache__/test_shm_csr.cpython-39-pytest-7.4.4.pyc has changed
--- a/tests/data/.~lock.handleiding activeren pas.docx#	Mon Jan 08 08:50:07 2024 +0000
+++ /dev/null	Thu Jan 01 00:00:00 1970 +0000
@@ -1,1 +0,0 @@
-Vorderman\, R.H.P. (MOLEPI) ,rhpvorderman,sasc-pc-6,21.02.2023 15:01,file:///home/rhpvorderman/.config/libreoffice/4;
\ No newline at end of file
Binary file tests/data/handleiding activeren pas.docx has changed
--- a/tests/test_shm_csr.py	Mon Jan 08 08:50:07 2024 +0000
+++ b/tests/test_shm_csr.py	Wed Jan 10 12:32:47 2024 +0000
@@ -49,8 +49,31 @@
         return files
     return ()
 
-@pytest.fixture(scope="module")
-def shm_csr_result():
+
+def run_shm_csr(
+    input=str(CONTROL_NWK377_PB_IGHC_MID1_40nt_2),
+    infile_name = "input_data",
+    functionality = "productive",
+    unique = "Sequence.ID",
+    naive_output = "no",
+    naive_output_ca = "None",
+    naive_output_cg = "None",
+    naive_output_cm = "None",
+    naive_output_ce = "None",
+    naive_output_all = "None",
+    naive_output_igm_naive = "None",
+    naive_output_igm_naive_memory = "None",
+    filter_unique = "remove",
+    filter_unique_count = '2',
+    class_filter = '70_70',
+    empty_region_filter = 'FR1',
+    # Skip baseline and changeo by default. These tools cannot be modified
+    # anyway and take most of the test time to execute. The environment
+    # variable can be set to "no" on the CI so the code path is tested
+    # at the time a PR is ready.
+    run_changeo = "yes" if os.environ.get("SHM_CSR_FAST") == "no" else "no",
+    run_baseline = "yes" if os.environ.get("SHM_CSR_FAST") == "no" else "no",
+):
     temp_dir = Path(tempfile.mkdtemp())
     tool_dir = temp_dir / "shm_csr"
     shutil.copytree(
@@ -62,29 +85,8 @@
     output_dir = temp_dir / "outputs"
     output_dir.mkdir(parents=True)
     wrapper = str(tool_dir / "wrapper.sh")
-    input = str(CONTROL_NWK377_PB_IGHC_MID1_40nt_2)
     out_files_path = output_dir / "results"
     out_file = out_files_path / "result.html"
-    infile_name = "input_data"
-    functionality = "productive"
-    unique = "Sequence.ID"
-    naive_output = "no"
-    naive_output_ca = "None"
-    naive_output_cg = "None"
-    naive_output_cm = "None"
-    naive_output_ce = "None"
-    naive_output_all = "None"
-    naive_output_igm_naive = "None"
-    naive_output_igm_naive_memory = "None"
-    filter_unique = "remove"
-    filter_unique_count = '2'
-    class_filter = '70_70'
-    empty_region_filter = 'FR1'
-    # Skip baseline and changeo by default. These tools cannot be modified
-    # anyway and take most of the test time to execute. The environment
-    # variable can be set to "no" on the CI so the code path is tested
-    # at the time a PR is ready.
-    fast = os.environ.get("SHM_CSR_FAST", "yes")
     cmd = [
         "bash",
         wrapper,
@@ -108,7 +110,8 @@
         filter_unique_count,
         class_filter,
         empty_region_filter,
-        fast
+        run_changeo,
+        run_baseline
     ]
     docker_cmd = ["docker", "run", "-v", f"{temp_dir}:{temp_dir}",
                   "--rm",  # Remove container after running
@@ -128,7 +131,12 @@
         with open(temp_dir / "stdout", "wt") as stdout_file:
             subprocess.run(docker_cmd, cwd=working_dir, stdout=stdout_file,
                            stderr=stderr_file, check=True)
-    yield Path(out_files_path)
+    return Path(out_files_path)
+
+
+@pytest.fixture(scope="module")
+def shm_csr_result():
+    yield run_shm_csr()
 
 
 def test_check_output(shm_csr_result):
@@ -157,3 +165,16 @@
                   ) as validate_h:
             for line in result_h:
                 assert line == validate_h.readline()
+
+
+def test_baseline_succeeds():
+    run_shm_csr(
+        functionality="unproductive",
+        empty_region_filter="None",
+        filter_unique="no",
+        unique="VGene,DGene,JGene,CDR3.IMGT.seq",
+        class_filter="101_101_IGA",
+        naive_output="yes",
+        run_baseline="yes",
+        run_changeo="yes",
+    )
Binary file time_ns has changed
--- a/wrapper.sh	Mon Jan 08 08:50:07 2024 +0000
+++ b/wrapper.sh	Wed Jan 10 12:32:47 2024 +0000
@@ -22,7 +22,8 @@
 filter_unique_count=${18}
 class_filter=${19}
 empty_region_filter=${20}
-fast=${21}
+run_changeo=${21}
+run_baseline=${22}
 BASENAME=$(basename "$title")
 # Cut off .txz or .tgz suffix and also replace spaces with underscores.
 NEW_IMGT_PREFIX="new_IMGT_${BASENAME%.*}"
@@ -410,7 +411,7 @@
   rm 1_Summary.txt
 }
 
-if [[ "$fast" == "no" ]] ; then
+if [[ "$run_baseline" == "yes" ]] ; then
 
     
 
@@ -524,7 +525,7 @@
 
 echo "</div>" >> $output #CSR tab end
 
-if [[ "$fast" == "no" ]] ; then
+if [[ "$run_changeo" == "yes" ]] ; then
 
 	echo "---------------- change-o MakeDB ----------------"