https://github.com/microbiomedata/metat
Metatranscriptomics workflow
https://github.com/microbiomedata/metat
metatranscriptomics transcriptomics workflow
Last synced: 3 months ago
JSON representation
Metatranscriptomics workflow
- Host: GitHub
- URL: https://github.com/microbiomedata/metat
- Owner: microbiomedata
- Created: 2020-11-02T20:42:32.000Z (over 5 years ago)
- Default Branch: main
- Last Pushed: 2025-03-25T16:43:37.000Z (10 months ago)
- Last Synced: 2025-03-25T17:45:10.354Z (10 months ago)
- Topics: metatranscriptomics, transcriptomics, workflow
- Language: WDL
- Homepage:
- Size: 22.2 MB
- Stars: 4
- Watchers: 4
- Forks: 2
- Open Issues: 6
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# metaT: The Metatranscriptome Workflow
## Summary
This workflow is designed to analyze metatranscriptomes.
All parts of this workflow are housed in their own repositories and imported via WDL v1.0 https importing.
The following repositories are used in this workflow:
- [metaT_ReadsQC](https://github.com/microbiomedata/metaT_ReadsQC)
- [metaT_Assembly](https://github.com/microbiomedata/metaT_Assembly)
- [mg_annotation](https://github.com/microbiomedata/mg_annotation)
- [metaT_ReadCounts](https://github.com/microbiomedata/metaT_ReadCounts)
## Version
0.0.6
## Third party tools and packages
To run this workflow you will need a Docker (Docker ≥ v2.1.0.3) instance and cromwell. All the third party tools are pulled from Dockerhub.
```
bbtools ≥ v38.94
Python ≥ v3.7.12
pandas ≥ v1.0.5 (python package)
gffutils ≥ v0.10.1 (python package)
```
## Databases
metaT uses the same database uses for metagenome annotation. See README [here](https://github.com/microbiomedata/mg_annotation) for required databases. For QC databases see [here](https://github.com/microbiomedata/ReadsQC.)
## Running workflow
### In a server with shifter
The submit script will request a node and launch the Cromwell. The Cromwell manages the workflow by using Shifter to run applications.
```
java -Dconfig.file=wdls/shifter.conf -jar /full/path/to/cromwell-XX.jar run -i input.json /full/path/to/wdls/metaT.wdl
```
## Docker images
- [microbiomedata/meta_t:0.0.5](https://hub.docker.com/r/microbiomedata/meta_t)
- [bryce911/bbtools:38.86](https://hub.docker.com/r/microbiomedata/bbtools)
## Inputs
```json
{
"metaT.input_files": ["./test_data/small_test/test_small_interleave.fastq.gz"],
"metaT.project_id":"nmdc:xxxxxxx",
"metaT.strand_type": "aRNA"
}
```
### Input option descriptions:
- `project_id`: A unique name for your project or sample.
- `input_file`: Full path to the fastq file. The file must be intereleaved paired end fastq.
- `input_fq1` and `input_fq2` if non-interleaved paired end fastqs
- `strand_type`: (optional) RNA strandedness, either left blank, `aRNA`, or `non_stranded_RNA`
## Outputs
All outputs can be found in the `outdir` folder. There are following subfolders:
- `outdir/annotation`: contains gff files from annotation run.
- `outdir/assembly`: contains FASTA files from assembly and BAM files where reads were mapped back to the contigs.
- `outdir/readMapping`: JSON files for sense and antisense that have records for feature, their annotations, read counts, ans associated statistics.
- `outdir/readsQC`: contains cleaned reads and a file with associated statistics.
# Output JSON
The output file is a JSON formatted file called `out.json` with JSON records that contains reads and information from annotation. An example JSON record:
```json
{
"featuretype": "CDS",
"seqid": "nmdc:xxxxxxx_001",
"id": "nmdc:xxxxxxx_001_1_588",
"source": "Prodigal v2.6.3_patched",
"start": 1,
"end": 588,
"length": 588,
"strand": "+",
"frame": "0",
"product": "hypothetical protein",
"product_source": "Hypo-rule applied",
"sense_read_count": 25,
"mean": 5.0,
"median": 3.0,
"stdev": 6.1,
"antisense_read_count": 28,
"meanA": 7.14,
"medianA": 7,
"stdevA": 5.7
}
```
## Test
To test the workflow, we have provided a small test dataset and a step by step guidance. See `test_data` folder.