Merge pull request #136 from nf-core/soft_matrix_support

Soft matrix support

.github/workflows/ci.yml

@@ -30,6 +30,7 @@ jobs:
           - "test"
           - "test_nogtf"
           - "test_affy"
+          - "test_soft"
     steps:
       - name: Check out pipeline code
         uses: actions/checkout@v3

CHANGELOG.md

@@ -11,6 +11,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - [[#129](https://github.com/nf-core/differentialabundance/pull/129)] - Module updates to fit with recent registry changes ([@pinin4fjords](https://github.com/pinin4fjords), review by [@maxulysse](https://github.com/maxulysse), [@adamrtalbot](https://github.com/adamrtalbot))
 - [[#130](https://github.com/nf-core/differentialabundance/pull/130)] - Document reasons for lack of differential expression ([@pinin4fjords](https://github.com/pinin4fjords), review by [@jfy133](https://github.com/jfy133))
 - [[#131](https://github.com/nf-core/differentialabundance/pull/131)] - Improve gtf to table configurability ([@pinin4fjords](https://github.com/pinin4fjords), review by [@WackerO](https://github.com/WackerO))
+- [# 136](https://github.com/nf-core/differentialabundance/pull/136)] - Added support for non-Affymetrix arrays via automatic download of SOFT matrices in GEO ([@azedinez](https://github.com/azedinez), review by [@pinin4fjords](https://github.com/pinin4fjords))
 - [[#137](https://github.com/nf-core/differentialabundance/pull/137)] - Add `--sizefactors_from_controls` and `--gene_id_col` for DESeq2 module to modules.config ([@WackerO](https://github.com/WackerO), review by [@pinin4fjords](https://github.com/pinin4fjords))
 - [[#145](https://github.com/nf-core/differentialabundance/pull/145)] - Template update for nf-core/tools v2.9 ([@nf-core-bot](https://github.com/nf-core-bot), review by [@pinin4fjords](https://github.com/pinin4fjords), [@WackerO](https://github.com/WackerO))
 

assets/differentialabundance_report.Rmd

@@ -210,7 +210,9 @@ if (! params$observations_name_col %in% colnames(observations)){
 
 if (! is.null(params$features)){
   features <- read_metadata(file.path(params$input_dir, params$features))
-  features <- features[,colnames(features) %in% simpleSplit(params$features_metadata_cols), drop = FALSE]
+  if (! is.null(params$features_metadata_cols)){
+    features <- features[,colnames(features) %in% simpleSplit(params$features_metadata_cols), drop = FALSE]
+  }
 }
 
 contrasts <- read_metadata(file.path(params$input_dir, params$contrasts_file))

conf/modules.config

@@ -100,6 +100,30 @@ process {
         ].join(' ').trim() }
     }
 
+    withName: GEOQUERY_GETGEO {
+        publishDir = [
+            [
+                path: { "${params.outdir}/tables/processed_abundance" },
+                mode: params.publish_dir_mode,
+                pattern: '*.matrix.tsv'
+            ],
+            [
+                path: { "${params.outdir}/tables/annotation" },
+                mode: params.publish_dir_mode,
+                pattern: '*.annotation.tsv'
+            ],
+            [
+                path: { "${params.outdir}/other/affy" },
+                mode: params.publish_dir_mode,
+                pattern: '*.{rds,sessionInfo.log}'
+            ]
+        ]
+        ext.prefix = { "normalised." }
+        ext.args = {
+            ((params.features_metadata_cols == null) ? '' : "--metacols \"${params.features_metadata_cols}\"")
+        }
+    }
+
     withName: DESEQ2_NORM {
         ext.prefix = 'all'
         publishDir = [

conf/soft.config

@@ -0,0 +1,46 @@
+/*
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Nextflow config file for running SOFT array file analysis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Defines settings specific to array analysis with SOFT files from GEO
+
+    Use as follows:
+        nextflow run nf-core/differentialabundance -profile soft,<docker/singularity> --outdir <OUTDIR>
+
+----------------------------------------------------------------------------------------
+*/
+
+params {
+
+    config_profile_name        = 'SOFT matrix track test profile'
+    config_profile_description = 'Minimal settings for test of the SOFT matrix track'
+
+    // Study
+    study_type = 'geo_soft_file'
+    study_abundance_type = 'intensities'
+
+    // Observations
+    observations_id_col = 'id'
+    observations_name_col = 'id'
+
+
+    // Features
+    features_id_col = 'ID'
+    features_metadata_cols = 'ID,ENTREZ_GENE_ID,Gene Symbol,Sequence Type'
+    features_name_col = 'Gene Symbol'
+
+
+    // Exploratory
+    exploratory_assay_names = 'normalised'
+    exploratory_final_assay = 'normalised'
+
+    // Differential options
+    differential_file_suffix         = ".limma.results.tsv"
+    differential_fc_column           = "logFC"
+    differential_pval_column         = "P.Value"
+    differential_qval_column         = "adj.P.Val"
+    differential_feature_id_column   = "probe_id"
+    differential_feature_name_column = "Symbol"
+
+}
+

conf/test_soft.config

@@ -0,0 +1,32 @@
+/*
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Nextflow config file for running minimal tests
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+    Defines input files and everything required to run a fast and simple
+    pipeline test with SOFT array files from GEO.
+
+    Use as follows:
+        nextflow run nf-core/differentialabundance -profile test_soft_array,<docker/singularity> --outdir <OUTDIR>
+
+----------------------------------------------------------------------------------------
+*/
+
+includeConfig 'soft.config'
+
+params {
+
+    config_profile_name        = 'SOFT matrix track test profile'
+    config_profile_description = 'Minimal settings for test of the SOFT matrix track'
+
+    // Limit resources so that this can run on GitHub Actions
+    max_cpus   = 2
+    max_memory = '6.GB'
+    max_time   = '6.h'
+
+    // Input
+    input = 'https://raw.githubusercontent.com/nf-core/test-datasets/differentialabundance/testdata/GSE50790.csv'
+    contrasts = 'https://raw.githubusercontent.com/nf-core/test-datasets/differentialabundance/testdata/GSE50790_contrasts.csv'
+    querygse = 'GSE50790'
+
+}
+

docs/usage.md

@@ -57,6 +57,22 @@ This is a numeric square matrix file, comma or tab-separated, with a column for
 
 This is an archive of CEL files as frequently found in GEO.
 
+### Use SOFT matrices
+
+Alternatively, the user may want to work with SOFT matrices. In this case, setting
+
+`--study_type geo_soft_file` and `--querygse [GSE study ID]`
+
+enables the pipeline to download normalised SOFT matrices automatically (note that even though Affymetrix arrays are also supported in the SOFT matrix track, it is recommended to work from CEL files in this case).
+
+As for other platforms You may subset the metadata features used in reporting etc. e.g. for GPL570 (Affymetrix Plus 2.0 arrays) this could be done with
+
+```
+--features_metadata_cols ID,Entrez_Gene_ID,Symbol,Definition
+```
+
+Full list of features metadata are available on GEO platform pages.
+
 ## Contrasts file
 
 ```bash

modules.json

@@ -40,6 +40,11 @@
                         "git_sha": "d0b4fc03af52a1cc8c6fb4493b921b57352b1dd8",
                         "installed_by": ["modules"]
                     },
+                    "geoquery/getgeo": {
+                        "branch": "master",
+                        "git_sha": "6814b0659c51e447684a58c2b834a9f3b530540d",
+                        "installed_by": ["modules"]
+                    },
                     "gsea/gsea": {
                         "branch": "master",
                         "git_sha": "911696ea0b62df80e900ef244d7867d177971f73",