Workflow: Run all steps

For convenience, seismic wf runs all steps of the main workflow:

Note

When you run the workflow with seismic wf, the steps Cluster, Fold, and Export do not run by default, but you can turn them on with options:

To cluster, use --max-clusters (-k) followed by the maximum number of clusters (in seismic wf, -k defaults to 0 – no clustering).
To fold, use --fold.
To export, use --export.

SEISMIC-RNA will process every input file from the point that it enters the workflow, following the gray arrays, until the end of the entire workflow. For example, for each file of read sequences that you input with -x, -y, or -z, it will

run Align on the FASTQ files of reads to yield alignment maps
run Relate on the alignment maps to yield Relate reports
run Mask on the Relate reports to yield Mask reports
(optionally) run Cluster on the Mask reports to make Cluster reports
run Table on the Relate, Mask, and (optionally) Cluster reports to yield table files
(optionally) run Fold on the Mask and (optionally) Cluster table files to yield predicted RNA secondary structure models
(optionally) run Export on the Relate, Mask, and (optionally) Cluster table files and (optionally) structure models to yield sample results
run Graph on the Relate, Mask, and (optionally) Cluster tables and (optionally) structure models to yield graph files

By contrast, for each Mask table file you input, SEISMIC-RNA will only

(optionally) run Fold on the table file to yield predicted RNA secondary structure models
(optionally) run Export on the table file and (optionally) structure models to yield sample results
run Graph on the table file and (optionally) structure models to yield graph files

Workflow: Input files

You can give any input file to seismic wf if you could give it to any individual command in the workflow.

The only required input file is of the reference sequences, which is used by the Align and Relate steps; see Relate input file: Reference sequences for more information. This file must be the first positional argument.

You can list additional input files (e.g. alignment maps, report files, table files) as positional arguments after the reference sequences, using any of the ways to list input files (see List Input Files), for example

seismic wf {refs.fa} sample1/refA.bam {out}/*/relate/refB/relate-report.json {out}/*/table/

where {refs.fa} is the path of your FASTA file of reference sequences and {out} is your top-level output directory.

You can put keyword input files (-x/-y/-z/-X/-Y/-Z) in any locations amid positional arguments, for example

seismic wf {refs.fa} -x sample4/ -y sample5.fq.gz -z sample6.fq.gz sample1/refA.bam {out}/*/relate/refB/relate-report.json {out}/*/table/

Workflow: Settings

You can use any option in seismic wf if you could use it in any individual command in the workflow. Additionally, --fold/--no-fold and --export/--no-export control whether the Fold and Export steps, respectively, will run in seismic wf (both are “no” by default). Options can be mixed in any locations amid positional arguments, for example

seismic wf --force {refs.fa} -x sample4/ --min-finfo-read 0.95 --min-ninfo-pos 100000 -k 3 {out}/*/table/ --fold

Workflow: Output files

The seismic wf command creates all output files that would have been created by running every step of the workflow individually. All output files go into the directory that you specify with --out-dir, which is ./out by default.

Workflow: Troubleshoot and optimize

Refer to the advice for the individual step you are troubleshooting/optimizing.