A machine learning approach to the identification of somatic recombination in immune cells 2462
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Abstract Description Recurrent recombination-based somatic driver events such as gene fusions are common in leukemia. One example is the IGHV gene, a key biomarker used for chronic lymphocytic leukemia (CLL). IGHV gene mutations or specific recombinations are linked with CLL prognosis. The clinical standard used in IGHV analysis is PCR-Sanger, however new bioinformatic tools for analyzing IGHV have emerged based on split-read analysis. We hypothesized that Machine Learning (ML) may improve IGHV analysis from sequencing data. Our objective was to classify reads as either IGHV or non-IGHV. Recently, studies have discovered that using synthetic data to supplement real data improves performance (De Melo et al., 2020). Thus, we have developed a synthetic dataset using a script that mimics B-cell IGHV recombination events. A natural language processing ML model was then trained on synthetic data, along with real data. BioAutoML was used to extract characteristics of the IGHV regions, thus enabling classification. DNABERT was used to identify specific V, D and J segments. A dataset of over 300,000 sequences was generated as input. The machine learning model was trained on synthetic data, and then tested on a set of synthetic data or real data, with F1-scores of 99.8% and 64.6% respectively. Although performance can be improved, our tools offer a potentially faster alternative to existing methods used to identify recombination events in immune cells with broad applications in cancer and immunology. Funding Sources This Project is supported by Research New Brunswick, the Beatrice Hunter Cancer Research Institute and the Atlantic Cancer Research Institute. Topic Categories Computational and Systems Immunology (COMP)
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.002 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 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 | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it