Deep learning models incorporating endogenous factors beyond DNA sequences improve the prediction accuracy of base editing outcomes
Bibliographic record
Abstract
Adenine base editors (ABEs) and cytosine base editors (CBEs) enable the single nucleotide editing of targeted DNA sites avoiding generation of double strand breaks, however, the genomic features that influence the outcomes of base editing in vivo still remain to be characterized. High-throughput datasets from lentiviral integrated libraries were used to investigate the sequence features affecting base editing outcomes, but the effects of endogenous factors beyond the DNA sequences are still largely unknown. Here the base editing outcomes of ABE and CBE were evaluated in mammalian cells for 5012 endogenous genomic sites and 11,868 genome-integrated target sequences, with 4654 genomic sites sharing the same target sequences. The comparative analyses revealed that the editing outcomes of ABE and CBE at endogenous sites were substantially different from those obtained using genome-integrated sequences. We found that the base editing efficiency at endogenous target sites of both ABE and CBE was influenced by endogenous factors, including epigenetic modifications and transcriptional activity. A deep-learning algorithm referred as BE_Endo, was developed based on the endogenous factors and sequence information from our genomic datasets, and it yielded unprecedented accuracy in predicting the base editing outcomes. These findings along with the developed computational algorithms may facilitate future application of BEs for scientific research and clinical gene therapy.
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 | 0.003 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".