Kallisto quantification

bin/salmon_tximport.r → bin/tximport.r

@@ -4,13 +4,14 @@ library(SummarizedExperiment)
 library(tximport)
 
 args = commandArgs(trailingOnly=TRUE)
-if (length(args) < 2) {
-    stop("Usage: salmon_tximport.r <coldata> <salmon_out>", call.=FALSE)
+if (length(args) < 3) {
+    stop("Usage: tximport.r <coldata> <quant_out> <type>", call.=FALSE)
 }
 
 coldata = args[1]
 path = args[2]
 sample_name = args[3]
+type = args[4]
 
 prefix = sample_name
 tx2gene = "salmon_tx2gene.tsv"
@@ -36,7 +37,7 @@ if (file.exists(coldata)) {
     coldata = data.frame(files = fns, names = names)
 }
 
-txi = tximport(fns, type = "salmon", txOut = TRUE)
+txi = tximport(fns, type = type, txOut = TRUE)
 rownames(coldata) = coldata[["names"]]
 extra = setdiff(rownames(txi[[1]]),  as.character(rowdata[["tx"]]))
 if (length(extra) > 0) {

conf/modules.config

@@ -1129,6 +1129,49 @@ if (!params.skip_multiqc) {
 // Salmon pseudo-alignment options
 //
 
+if (!params.skip_pseudo_alignment && params.pseudo_aligner == 'kallisto') {
+    process {
+        withName: '.*:QUANTIFY_KALLISTO:KALLISTO_QUANT' {
+            ext.fragment_len = { meta.fragment_len ?: params.kallisto_default_fraglen }
+            ext.sd = { meta.fragment_len ?: params.kallisto_default_fraglen_se }
+            ext.args   = params.extra_kallisto_quant_args ?: ''
+            publishDir = [
+                path: { "${params.outdir}/${params.pseudo_aligner}" },
+                mode: params.publish_dir_mode,
+                saveAs: { filename -> filename.equals('versions.yml') || filename.endsWith('_meta_info.json') ? null : filename }
+            ]
+        }
+
+        withName: '.*:QUANTIFY_KALLISTO:SALMON_TX2GENE' {
+            publishDir = [
+                path: { "${params.outdir}/${params.pseudo_aligner}" },
+                mode: params.publish_dir_mode,
+                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            ]
+        }
+
+        withName: '.*:QUANTIFY_KALLISTO:SALMON_TXIMPORT' {
+            publishDir = [
+                path: { "${params.outdir}/${params.pseudo_aligner}" },
+                mode: params.publish_dir_mode,
+                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            ]
+        }
+
+        withName: '.*:QUANTIFY_KALLISTO:SALMON_SE_.*' {
+            publishDir = [
+                path: { "${params.outdir}/${params.pseudo_aligner}" },
+                mode: params.publish_dir_mode,
+                saveAs: { filename -> filename.equals('versions.yml') ? null : filename }
+            ]
+        }
+    }
+}
+
+//
+// Salmon pseudo-alignment options
+//
+
 if (!params.skip_pseudo_alignment && params.pseudo_aligner == 'salmon') {
     process {
         withName: '.*:QUANTIFY_SALMON:SALMON_QUANT' {

modules.json

@@ -65,6 +65,16 @@
                         "git_sha": "735e1e04e7e01751d2d6e97055bbdb6f70683cc1",
                         "installed_by": ["modules"]
                     },
+                    "kallisto/index": {
+                        "branch": "master",
+                        "git_sha": "8fc1d24c710ebe1d5de0f2447ec9439fd3d9d66a",
+                        "installed_by": ["modules"]
+                    },
+                    "kallisto/quant": {
+                        "branch": "master",
+                        "git_sha": "8fc1d24c710ebe1d5de0f2447ec9439fd3d9d66a",
+                        "installed_by": ["modules"]
+                    },
                     "picard/markduplicates": {
                         "branch": "master",
                         "git_sha": "2ee934606f1fdf7fc1cb05d6e8abc13bec8ab448",

modules/local/salmon_tximport/main.nf → modules/local/tximport/main.nf

@@ -24,7 +24,7 @@ process SALMON_TXIMPORT {
 
     script: // This script is bundled with the pipeline, in nf-core/rnaseq/bin/
     """
-    salmon_tximport.r \\
+    tximport.r \\
         NULL \\
         salmon \\
         salmon.merged

nextflow.config

@@ -66,6 +66,9 @@ params {
     min_mapped_reads           = 5
     extra_star_align_args      = null
     extra_salmon_quant_args    = null
+    extra_kallisto_quant_args  = null
+    kallisto_default_fraglen   = 150
+    kallisto_default_fraglen_se  = 75
     save_merged_fastq          = false
     save_unaligned             = false
     save_align_intermeds       = false

nextflow_schema.json

@@ -149,6 +149,13 @@
                     "fa_icon": "fas fa-bezier-curve",
                     "description": "Path to directory or tar.gz archive for pre-built Salmon index."
                 },
+                "kallisto_index": {
+                    "type": "string",
+                    "format": "path",
+                    "exists": true,
+                    "fa_icon": "fas fa-bezier-curve",
+                    "description": "Path to directory or tar.gz archive for pre-built Kallisto index."
+                },
                 "hisat2_build_memory": {
                     "type": "string",
                     "default": "200.GB",
@@ -352,7 +359,7 @@
                     "type": "string",
                     "description": "Specifies the pseudo aligner to use - available options are 'salmon'. Runs in addition to '--aligner'.",
                     "fa_icon": "fas fa-hamburger",
-                    "enum": ["salmon"]
+                    "enum": ["salmon", "kallisto"]
                 },
                 "bam_csi_index": {
                     "type": "boolean",
@@ -396,6 +403,23 @@
                     "type": "string",
                     "description": "Extra arguments to pass to Salmon quant command in addition to defaults defined by the pipeline.",
                     "fa_icon": "fas fa-plus"
+                },
+                "extra_kallisto_quant_args": {
+                    "type": "string",
+                    "description": "Extra arguments to pass to Kallisto quant command in addition to defaults defined by the pipeline.",
+                    "fa_icon": "fas fa-plus"
+                },
+                "kallisto_default_fraglen": {
+                    "type": "integer",
+		    "description": "In single-end mode Kallisto requires an estimated fragment length. Specify a default value for that here. TODO: use existing RSeQC results to do this dynamically.",
+                    "default": 150,
+                    "fa_icon": "fas fa-ruler-horizontal"
+                },
+                "kallisto_default_fraglen_se": {
+                    "type": "integer",
+		    "description": "In single-end mode Kallisto requires an estimated standard error for fragment length. Specify a default value for that here. TODO: use existing RSeQC results to do this dynamically.",
+                    "default": 75,
+                    "fa_icon": "fas fa-sort-amount-up-alt"
                 }
             }
         },