Published Paper
Download Model
GitHub Repository 1 (legacy)
GitHub Repository 2 (legacy)
The links marked legacy point to the original published source code. For the model as this web server runs it, use the Download Model link above.
DeepSpCas9 designs guide RNAs and predicts the efficiency of CRISPR–Cas9 (SpCas9) with conventional scaffold at endogenous target sites.
It supports the input of target sequences (at least 30bp) via FASTA format or a gene symbol in the Human Gene List, and seven further formats: NMID, Ensembl ID, HGNC ID, a genomic position, an unedited/edited sequence pair, ClinVar ID and COSMIC ID.
Homology-directed repair (HDR) donor templates. When an input format describes an intended edit — an unedited/edited sequence pair, ClinVar ID or COSMIC ID — the results also carry what each guide needs for HDR: the cut site, the cut-to-edit distance, the donor template in the chosen type and orientation(s) (default: a single-stranded oligo on the sense strand, 100 nt homology arms each side), the realised arm lengths, an optional PAM-blocking mutation, and the candidate's rank among the target's guides. There is no HDR mode switch: the input format decides.
Note: DeepSpCas9variants provides guide RNA designs and efficiency predictions with optimized scaffold.
