ChromeCRISPR - A High Efficacy Hybrid Machine Learning Model for CRISPR/Cas On-Target Predictions
Bibliographic record
Abstract
Abstract Genome editing has the potential to treat genetic disorders at the source. This can be achieved by modifying the defective DNA through the intentional insertion, deletion, or substitution of genomic content. Among all genome editing technologies, CRISPR/Cas (Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-associated protein) is considered the gold standard. CRISPR/Cas uses a single guide RNA (sgRNA) to direct the Cas nuclease to a target DNA region. Due to the ease at creating small RNA molecules, it is possible to have the CRISPR/Cas complex target any arbitrary DNA sequence, thus making it a versatile tool. The efficacy of the complex is dependent on the ability of the sgRNA to bind to a complementary DNA sequence, which varies based on the sequence. Thus, a major challenge is finding sgRNA sequences that have good efficacy. This is where computational models can aid scientists: by predicting the activity of sgRNAs to help narrow the search space of finding the optimal sgRNA. We have used a large new dataset to build and compare the ability of several different machine learning architectures’ ability to predict on-target CRISPR/Cas activity. Additionally, we explored how adding GC content affects our sgRNA activity predictions. Our novel hybrid model, ChromeCRISPR, combines the strengths of Convolutional Neural Networks (CNN) with Recurrent Neural Network (RNN) models, has outperformed state-of-the-art models, including DeepHF and AttCrispr, establishing a new benchmark for predictive accuracy in CRISPR/Cas9 efficacy predictions.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.003 | 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".