Pangolin RNA Splicing Prediction
Pangolin is a deep learning model to predict splice site strength and the impact of genetic variants on RNA splicing in multiple tissues.
Pangolin is a deep learning model that predicts RNA splicing from DNA sequence with high accuracy. It outperforms state-of-the-art methods in predicting splice site strength across multiple tissues (heart, liver, brain, and testis). By training on data from four species (human, rhesus macaque, rat, and mouse), Pangolin effectively predicts the impact of common, rare, and lineage-specific genetic variants on RNA splicing. It shows remarkable potential for identifying pathogenic, loss-of-function mutations, especially those that are not missense or nonsense, making it a valuable tool for interpreting disease-causing variants.
Estimated resource cost: 0.000248
Categories: Multi-omics
Tags: Genomics
Key Capabilities
- Predicts splice site probability and usage for four organs: heart, liver, brain, and testis.
- Trained on sequence and RNA splicing data from four different species: human, rhesus macaque, mouse, and rat.
- Outperforms existing methods like SpliceAI, MMSplice, and HAL on splice site prediction.
- Accurately predicts the effects of rare and common genetic variants on RNA splicing.
- Identifies loss-of-function (LOF) mutations with high precision and recall, aiding in the identification of pathogenic variants.
- Models epistatic effects, enabling accurate prediction for combinations of genetic variants.
Runtime Statistics
| Metric | Value |
|---|---|
| runtime_mean | 9 |
| runtime_median | 6 |
| runtime_std | 9 |
| runtime_90th_percentile | 29 |
| runtime_max | 33 |
Similar Tools
- DEAnalysis
- Interleaved FASTQ Splitter
- Transcript Assembly
Ready to submit your job?
Review your configuration, then confirm the estimated credit cost before you run the job. Note that credit estimates are not guaranteed and runtime can vary depending on inputs and settings.
Estimated Credits: 0.000248
Invite-only, limited-time access. Please contact ztang@getantibody.com.
Invite-only, limited-time access. Please contact ztang@getantibody.com.