Feature: co-assemblies submission with --mode metagenomic_assembly

CHANGELOG.md

@@ -5,6 +5,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## v1.0.1dev - [date]
 
+### `Added`
+
+- [#66](https://github.com/nf-core/seqsubmit/pull/66)
+  - Added co-assembly submission to `ASSEMBLYSUBMIT`: a samplesheet row can list several semicolon-separated run accessions (and matching read files for coverage calculation)
+  - When the source runs belong to different ENA samples, a virtual sample with a `composed of` attribute is registered and the co-assembly is submitted under it
+  - Optional `tax_id` samplesheet column to set the taxon of the virtual co-assembly sample when the source samples have different taxa
+
 ### `Fixed`
 
 - [#117](https://github.com/nf-core/seqsubmit/pull/117) and [#119](https://github.com/nf-core/seqsubmit/pull/119)

README.md

@@ -74,8 +74,11 @@ Example:
 id,fasta,fastq_1,fastq_2,coverage,run_accession,assembler,assembler_version,assembly_type
 assembly_1,data/contigs_1.fasta.gz,data/reads_1.fastq.gz,data/reads_2.fastq.gz,,ERR011322,SPAdes,3.15.5,metagenome
 assembly_2,data/contigs_2.fasta.gz,,,42.7,ERR011323,MEGAHIT,1.2.9,metatranscriptome
+coassembly_1,data/contigs_3.fasta.gz,data/run1_R1.fastq.gz;data/run2_R1.fastq.gz,data/run1_R2.fastq.gz;data/run2_R2.fastq.gz,,ERR011324;ERR011325,MEGAHIT,1.2.9,
 ```
 
+Co-assemblies are supported: list all source runs in `run_accession` and their read files in `fastq_1`/`fastq_2`, separated by semicolons.
+
 See the [`metagenomic_assemblies` mode section](docs/usage.md#samplesheet-input-1) of the usage documentation for more details.
 
 ### `mags` and `bins` modes

assets/schema_input_assembly.json

@@ -26,9 +26,7 @@
                 "anyOf": [
                     {
                         "type": "string",
-                        "format": "file-path",
-                        "exists": true,
-                        "pattern": "^\\S+\\.(fq|fastq)(\\.gz)?$"
+                        "pattern": "^\\S+\\.(fq|fastq)(\\.gz)?(;\\S+\\.(fq|fastq)(\\.gz)?)*$"
                     },
                     {
                         "type": "string",
@@ -37,15 +35,13 @@
                 ],
                 "default": null,
                 "errorMessage": "FASTQ file must have extension '.fq' or '.fastq' (optionally gzipped)",
-                "description": "Path to single-end/paired-end forward reads in FASTQ format (optionally gzipped). Required if coverage is not provided"
+                "description": "Path to single-end/paired-end forward reads in FASTQ format (optionally gzipped). Required if coverage is not provided. For co-assemblies, semicolon-separated paths (e.g. run1_R1.fq.gz;run2_R1.fq.gz)"
             },
             "fastq_2": {
                 "anyOf": [
                     {
                         "type": "string",
-                        "format": "file-path",
-                        "exists": true,
-                        "pattern": "^\\S+\\.(fq|fastq)(\\.gz)?$"
+                        "pattern": "^\\S+\\.(fq|fastq)(\\.gz)?(;\\S+\\.(fq|fastq)(\\.gz)?)*$"
                     },
                     {
                         "type": "string",
@@ -54,7 +50,7 @@
                 ],
                 "default": null,
                 "errorMessage": "FASTQ file for reverse reads must have extension '.fq' or '.fastq' (optionally gzipped)",
-                "description": "Path to reverse reads in FASTQ format for paired-end data (optionally gzipped). Leave empty for single-end reads"
+                "description": "Path to reverse reads in FASTQ format for paired-end data (optionally gzipped). Leave empty for single-end reads. For co-assemblies, use semicolon-separated paths (e.g. run1_R2.fq.gz;run2_R2.fq.gz)"
             },
             "coverage": {
                 "anyOf": [
@@ -76,7 +72,7 @@
                 "type": "string",
                 "pattern": "^\\S+$",
                 "errorMessage": "Accession must be provided and cannot contain spaces",
-                "description": "ENA run accession for the reads used to generate the metagenomic assembly. Reads must already be submitted to ENA",
+                "description": "ENA run accession(s) for the reads used to generate the metagenomic assembly. Reads must already be submitted to ENA. For co-assemblies, use semicolon-separated values (e.g. ERR000001;ERR000002)",
                 "meta": ["run_accession"]
             },
             "assembler": {
@@ -100,6 +96,22 @@
                 "description": "Assembly type: metagenome or metatranscriptome. Written to the ENA manifest as ASSEMBLY_TYPE 'primary metagenome' or 'metatranscriptome'. Default: metagenome",
                 "default": "metagenome",
                 "meta": ["assembly_type"]
+            },
+            "tax_id": {
+                "anyOf": [
+                    {
+                        "type": "integer",
+                        "exclusiveMinimum": 0
+                    },
+                    {
+                        "type": "string",
+                        "maxLength": 0
+                    }
+                ],
+                "default": null,
+                "errorMessage": "Taxon ID must be a positive integer or empty",
+                "description": "NCBI taxon ID for the virtual sample registered for a co-assembly (optional). Provide when the source samples have different taxon IDs. Ignored for assemblies that are not co-assemblies of several samples",
+                "meta": ["tax_id"]
             }
         },
         "required": ["id", "fasta", "run_accession", "assembler", "assembler_version"],