Predicting CRISPR-Cas9 off-target effects in human primary cells using bidirectional LSTM with BERT embedding
Bibliographic record
Abstract
Abstract Motivation Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR)-Cas9 system is a ground-breaking genome editing tool, which has revolutionized cell and gene therapies. One of the essential components involved in this system that ensures its success is the design of an optimal single-guide RNA (sgRNA) with high on-target cleavage efficiency and low off-target effects. This is challenging as many conditions need to be considered, and empirically testing every design is time-consuming and costly. In silico prediction using machine learning models provides high-performance alternatives. Results We present CrisprBERT, a deep learning model incorporating a Bidirectional Encoder Representations from Transformers (BERT) architecture to provide a high-dimensional embedding for paired sgRNA and DNA sequences and Bidirectional Long Short-term Memory networks for learning, to predict the off-target effects of sgRNAs utilizing only the sgRNAs and their paired DNA sequences. We proposed doublet stack encoding to capture the local energy configuration of the Cas9 binding and applied the BERT model to learn the contextual embedding of the doublet pairs. Our results showed that the new model achieved better performance than state-of-the-art deep learning models regarding single split and leave-one-sgRNA-out cross-validations as well as independent testing. Availability and implementation The CrisprBERT is available at GitHub: https://github.com/OSsari/CrisprBERT.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".