A Comparison of Machine Learning Models for Predicting CRISPR/Cas On-target Efficacy
Bibliographic record
Abstract
CRISPR/Cas (Clustered Regularly Interspaced Short Palindromic Repeats and CRISPR-associated protein) is a powerful technology that can precisely modify DNA, enabling the treatment of several genetic disorders. The CRISPR/Cas system is comprised of a nuclease, which induces the modification, and a sgRNA (single guide RNA), which targets the nuclease to a precise location in the DNA. Designing sgRNAs is time-consuming and resource-intensive, thus computational tools are used to screen sgRNAs for their on-target efficacy. However, to date, models for predicting on-target efficacy have been restricted in complexity due to data limitations. Recently, a large on-target dataset has been published, relieving some of the restraints towards creating larger computational models. Herein, we present a comparison of several types of deep learning models, including CNNs (Convolutional Neural Networks) and RNNs (Recurrent Neural Networks), using the new dataset to predict the on-target efficacy of sgRNAs. After determining which general model performs best on this dataset, we assessed the impact of adding an important biological feature, the GC content, and compared the models to a state-of-the-art competitor, DeepHF [1].
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 | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.003 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".