---
name: bio-sashimi-plots
source: https://app.decimal.ai/s/bio-sashimi-plots@1/SKILL.md
source_sha256: 347367c7adef
---

## Version Compatibility

Reference examples tested with: ggplot2 3.5+, pandas 2.2+

Before using code patterns, verify installed versions match. If versions differ:
- Python: `pip show <package>` then `help(module.function)` to check signatures
- CLI: `<tool> --version` then `<tool> --help` to confirm flags

If code throws ImportError, AttributeError, or TypeError, introspect the installed
package and adapt the example to match the actual API rather than retrying.

# Sashimi Plot Visualization

Create sashimi plots to visualize splicing events with read coverage and junction counts.

## ggsashimi Usage

**Goal:** Generate sashimi plots showing read coverage and junction counts for a genomic region.

**Approach:** Define sample groupings in a TSV file, then run ggsashimi with genomic coordinates and annotation.

**"Visualize a splicing event"** -> Plot RNA-seq coverage tracks with splice junction arcs grouped by condition.
- Python/CLI: `ggsashimi.py` (ggsashimi)
- CLI: `rmats2sashimiplot` (rMATS-specific)

```python
import subprocess
import pandas as pd

# Create sample grouping file (TSV: path, group, color)
groups = pd.DataFrame({
    'bam': ['sample1.bam', 'sample2.bam', 'sample3.bam', 'sample4.bam'],
    'group': ['control', 'control', 'treatment', 'treatment'],
    'color': ['#1f77b4', '#1f77b4', '#ff7f0e', '#ff7f0e']
})
groups.to_csv('sashimi_groups.tsv', sep='\t', index=False, header=False)

# Basic sashimi plot for a region
subprocess.run([
    'ggsashimi.py',
    '-b', 'sashimi_groups.tsv',
    '-c', 'chr1:1000000-1010000',  # Genomic coordinates
    '-o', 'sashimi_output',
    '-M', '10',  # Minimum junction reads to show
    '--alpha', '0.25',  # Coverage transparency
    '--height', '3',
    '--width', '8',
    '-g', 'annotation.gtf'
], check=True)
```

## Batch Plotting Significant Events

**Goal:** Automatically generate sashimi plots for all significant differential splicing events.

**Approach:** Load rMATS results, filter for significant events, extract flanking coordinates, and iterate ggsashimi over each event.

```python
import subprocess
import pandas as pd

# Load differential splicing results
diff_results = pd.read_csv('rmats_output/SE.MATS.JC.txt', sep='\t')
significant = diff_results[
    (diff_results['FDR'] < 0.05) &
    (diff_results['IncLevelDifference'].abs() > 0.1)
]

# Generate plots for top events
for idx, event in significant.head(20).iterrows():
    chrom = event['chr']
    # Extend region around the exon
    start = event['upstreamES'] - 500
    end = event['downstreamEE'] + 500
    region = f'{chrom}:{start}-{end}'
    gene = event['geneSymbol']

    subprocess.run([
        'ggsashimi.py',
        '-b', 'sashimi_groups.tsv',
        '-c', region,
        '-o', f'sashimi_plots/{gene}_{chrom}_{start}',
        '-M', '5',
        '--shrink',  # Shrink introns for better visualization
        '-g', 'annotation.gtf',
        '--fix-y-scale'  # Same y-axis across groups
    ], check=True)
```

## rmats2sashimiplot

**Goal:** Create sashimi plots directly from rMATS differential splicing output.

**Approach:** Point rmats2sashimiplot at rMATS result files and BAM groups with condition labels.

```bash
# For rMATS output specifically
rmats2sashimiplot \
    --b1 sample1.bam,sample2.bam \
    --b2 sample3.bam,sample4.bam \
    -t SE \
    -e rmats_output/SE.MATS.JC.txt \
    --l1 Control \
    --l2 Treatment \
    -o sashimi_rmats \
    --exon_s 1 \
    --intron_s 5
```

## Customization Options

**Goal:** Fine-tune sashimi plot appearance for publication-quality figures.

**Approach:** Adjust ggsashimi visual parameters including intron shrinking, y-axis scaling, aggregation mode, and output format.

```python
# Advanced ggsashimi options
subprocess.run([
    'ggsashimi.py',
    '-b', 'sashimi_groups.tsv',
    '-c', 'chr1:1000000-1010000',
    '-o', 'custom_sashimi',
    '-g', 'annotation.gtf',

    # Visual options
    '-M', '10',           # Min junction reads
    '--alpha', '0.25',    # Coverage alpha
    '--height', '3',      # Plot height per track
    '--width', '10',      # Plot width
    '--base-size', '14',  # Font size

    # Layout options
    '--shrink',           # Shrink introns
    '--fix-y-scale',      # Same y-axis
    '-A', 'mean',         # Aggregate: mean, median, or none

    # Annotation options
    '--gtf-filter', 'protein_coding',  # Filter GTF features

    # Output format
    '-F', 'pdf'           # pdf, png, svg, eps
], check=True)
```

## Best Practices

| Tip | Rationale |
|-----|-----------|
| Use `--shrink` for large introns | Keeps exons visible |
| Set `--fix-y-scale` for comparisons | Fair visual comparison |
| Aggregate replicates with `-A mean` | Reduces clutter |
| Limit to 3-4 groups | More groups become hard to read |
| Include flanking exons | Show full splicing context |

## Troubleshooting

| Issue | Solution |
|-------|----------|
| No junctions shown | Lower `-M` threshold |
| Plot too crowded | Use `--shrink`, reduce samples |
| Annotation missing | Check GTF format, gene name field |
| Memory issues | Plot smaller regions |

## Related Skills

- differential-splicing - Identify events to plot
- splicing-quantification - Context for PSI values
- data-visualization/ggplot2-fundamentals - Further customization