tiger.md

## TIGER Tool for Cas13 Efficacy Prediction

Welcome to TIGER! This online tool accompanies our *Nature Biotechnology* article.
TIGER's ability to make accurate on- and off-target predictions enables users to 1) design highly effective gRNAs and 2) precisely modulate transcript expression by engineered gRNA-target mismatches. 

If you use TIGER, please consider citing our study:
> Wessels, H.-H., Stirn, A., Méndez-Mancilla, A., Kim, E. J., Hart, S. K., Knowles, D. A., & Sanjana, N. E. (2023). Prediction of on-target and off-target activity of CRISPR–Cas13d guide RNAs using deep learning. Nature Biotechnology. https://doi.org/10.1038/s41587-023-01830-8


Please note that this precompiled, online tool differs from the manuscript slightly.
First, this version of TIGER predicts using just target and guide sequence (see Figure 3c). Second, we map TIGER's predictions to the unit interval to make estimates more interpretable: a Guide Score close to 1 corresponds to high gRNA activity (i.e. desirable for on-target guides).
A Guide Score near 0 denotes no/minimal activity (i.e. desirable for off-target effects).
This transformation is monotonic and therefore preserves Spearman, AUROC, and AUPRC performance.
These estimates (transformations of log-fold-change predictions from TIGER) appear in the "Guide Score" column of this online tool’s output.

### Using the TIGER Online Tool

The tool supports two methods for transcript entry:
1) Manual entry of a single transcript
2) Uploading a FASTA file that can contain one or more transcripts. Each transcript **must** have a unique ID.

The tool has three run modes:
- Report all on-target gRNAs for each provided transcript. This mode does not support off-target identification.
- Report the top 10 most active, on-target gRNAs for each provided transcript. This mode allows for the optional identification of off-target effects.
- Report the top 10 most active, on-target gRNAs for each provided transcript and their titration candidates (all possible single mismatches). Larger guide scores correspond to more transcript knockdown. This mode also does not support off-target identification.

The tool uses version 19 of Gencode (protein-coding and lncRNA) to identify off-target candidates.

### Future Development Plans

- Off-target scanning speed improvements
- Off-target scanning for titration (engineered mismatch) mode
- Allow users to select more than the top ten guides per transcript
- Incorporate non-scalar features (target accessibility, hybridization energies, etc...)

To report bugs or to request additional features, please click the "Community" button in the top right corner of this screen and start a new discussion.
Alternatively, please email [Andrew Stirn](mailto:andrew.stirn@cs.columbia.edu).