Command-line interface¶
Installing the package puts one command on your PATH, splicescope, with three
subcommands. The help text below is generated from the parser when the site is built,
so it always matches the installed version.
| subcommand | what it does |
|---|---|
simulate |
write a synthetic dataset with known ground truth |
run |
the whole pipeline on a directory of STAR junction tables |
consequence |
protein-consequence prediction for events produced by any tool |
splicescope¶
usage: splicescope [-h] [--version] {simulate,run,consequence} ...
Command-line interface for splicescope.
positional arguments:
{simulate,run,consequence}
simulate write a synthetic ground-truth dataset
run end-to-end: annotate -> quantify -> diff -> cryptic ML
consequence predict frame / PTC / NMD effects for cassette exons or
splice-site shifts
options:
-h, --help show this help message and exit
--version show program's version number and exit
splicescope simulate¶
Writes STAR SJ.out.tab files, a GTF annotation with coding sequences, groups.tsv, an
indexed genome FASTA and truth.tsv, which lists the injected events so that recall can
be measured. Nothing is downloaded.
usage: splicescope simulate [-h] --outdir OUTDIR [--genes GENES]
[--replicates REPLICATES] [--alt-ss ALT_SS] [--mxe MXE]
[--seed SEED]
options:
-h, --help show this help message and exit
--outdir OUTDIR
--genes GENES
--replicates REPLICATES
--alt-ss ALT_SS fraction of genes with an A5SS/A3SS event
--mxe MXE fraction of genes with an MXE event
--seed SEED
splicescope run¶
Annotates, quantifies, detects events, tests for differential splicing and, when
--genome is given, predicts protein consequences. See Quick start for
the files it writes.
usage: splicescope run [-h] --sj-dir SJ_DIR --gtf GTF --groups GROUPS --outdir OUTDIR
[--gene-sets GENE_SETS] [--labels LABELS] [--genome GENOME]
[--min-reads MIN_READS] [--max-exon MAX_EXON]
[--max-mxe-candidates MAX_MXE_CANDIDATES] [--seed SEED]
options:
-h, --help show this help message and exit
--sj-dir SJ_DIR directory of STAR *.SJ.out.tab files
--gtf GTF
--groups GROUPS TSV with columns sample,condition
--outdir OUTDIR
--gene-sets GENE_SETS
optional GMT file for pathway enrichment
--labels LABELS optional TSV of curated cryptic labels, to train and score the
classifier: columns chrom, start, end, strand,
is_cryptic_truth (0/1)
--genome GENOME indexed genome FASTA (.fai required); enables protein-
consequence prediction
--min-reads MIN_READS
--max-exon MAX_EXON longest candidate exon considered for mutually-exclusive-exon
events
--max-mxe-candidates MAX_MXE_CANDIDATES
most candidate exons one donor/acceptor anchor may hold before
it is treated as noise and skipped (pair enumeration is
quadratic in this)
--seed SEED
The groups file is a tab-separated table with the columns sample and condition, and
exactly two conditions. Sample names come from the junction file names with a trailing
SJ.out removed: A1_SJ.out.tab, A1.SJ.out.tab, A1SJ.out.tab and A1.tab all give
A1. The run stops with an error if no sample name is shared between the two.
splicescope consequence¶
Runs only the consequence layer, on candidates produced elsewhere (for example an rMATS
or LeafCutter run). Cassette exons are given as exon intervals (exon_start,
exon_end); novel donors and acceptors as junctions (start, end). The genome needs
its samtools faidx index next to it.
usage: splicescope consequence [-h] --events EVENTS --gtf GTF --genome GENOME --out
OUT [--mode {auto,exon,junction}]
options:
-h, --help show this help message and exit
--events EVENTS TSV of cassette exons (chrom, strand, exon_start, exon_end) or
of junctions (chrom, strand, start, end)
--gtf GTF
--genome GENOME indexed genome FASTA (.fai required)
--out OUT
--mode {auto,exon,junction}
'exon' reads exon_start/exon_end as a cassette exon;
'junction' reads start/end as an intron and resolves its novel
splice site against the annotation; 'auto' (default) picks by
which columns are present
The consequence classes, the frame arithmetic and the NMD rule are defined in Methods, section 8b.