roughing out a sample analysis workflow

workflows-htc/bioinformatics-samples/README.md

@@ -0,0 +1,10 @@
+# Submitting a Workflow that Analyzes Sequencing Samples
+
+This assumes that for what you want to process on CHTC, 
+the samples are each processed individually. 
+
+Assumes files are in /staging
+
+Make a log dir
+
+this will submit 3 jobs

workflows-htc/bioinformatics-samples/sample-analysis.sh

@@ -0,0 +1,19 @@
+#!/bin/bash
+
+# Set the sample id: 
+SAMPLEID=$1
+# if you used the option with a lane argument
+# uncomment to set the lane value: 
+# LANE=$2
+
+# build up the filenames - edit this to match
+# the filenames you have! 
+READ1=${SAMPLEID}_R1_001.fastq
+READ2=${SAMPLEID}_R2_001.fastq
+
+# insert any needed software setup here, if relevant
+
+# run your program 
+# (replace 'head' with your analysis pipeline)
+head ${READ1}
+head ${READ2}

workflows-htc/bioinformatics-samples/sample-analysis.sub

@@ -0,0 +1,24 @@
+universe = vanilla
+
+executable = sample-analysis.sh
+arguments = $(SampleID)
+# alternate form, for both samples/lanes
+# arguments = $(SampleID) $(Lane)
+
+# transfer_input_files = /path/to/file1,/path/to/file2,etc
+should_transfer_files = YES
+when_to_transfer_output = ON_EXIT
+
+# adjust these based on how much computing power 
+# you need to analyze ONE set of paired end reads
+request_cpus = 1
+request_memory = 1GB
+request_disk = 1GB
+
+log = logs/$(Cluster).log
+error = logs/$(Cluster)_$(SampleID).err
+output = logs/$(Cluster)_$(SampleID).out
+
+queue SampleID from samples.txt
+# alternate form, for both samples/lanes
+# queue SampleID,Lane from samples.txt

workflows-htc/bioinformatics-samples/samples.txt

@@ -0,0 +1,3 @@
+CS001
+CS006
+CS905