SNooPer: a machine learning-based method for somatic variant identification from low-pass next-generation sequencing
Bibliographic record
Abstract
BACKGROUND: Next-generation sequencing (NGS) allows unbiased, in-depth interrogation of cancer genomes. Many somatic variant callers have been developed yet accurate ascertainment of somatic variants remains a considerable challenge as evidenced by the varying mutation call rates and low concordance among callers. Statistical model-based algorithms that are currently available perform well under ideal scenarios, such as high sequencing depth, homogeneous tumor samples, high somatic variant allele frequency (VAF), but show limited performance with sub-optimal data such as low-pass whole-exome/genome sequencing data. While the goal of any cancer sequencing project is to identify a relevant, and limited, set of somatic variants for further sequence/functional validation, the inherently complex nature of cancer genomes combined with technical issues directly related to sequencing and alignment can affect either the specificity and/or sensitivity of most callers. RESULTS: For these reasons, we developed SNooPer, a versatile machine learning approach that uses Random Forest classification models to accurately call somatic variants in low-depth sequencing data. SNooPer uses a subset of variant positions from the sequencing output for which the class, true variation or sequencing error, is known to train the data-specific model. Here, using a real dataset of 40 childhood acute lymphoblastic leukemia patients, we show how the SNooPer algorithm is not affected by low coverage or low VAFs, and can be used to reduce overall sequencing costs while maintaining high specificity and sensitivity to somatic variant calling. When compared to three benchmarked somatic callers, SNooPer demonstrated the best overall performance. CONCLUSIONS: While the goal of any cancer sequencing project is to identify a relevant, and limited, set of somatic variants for further sequence/functional validation, the inherently complex nature of cancer genomes combined with technical issues directly related to sequencing and alignment can affect either the specificity and/or sensitivity of most callers. The flexibility of SNooPer's random forest protects against technical bias and systematic errors, and is appealing in that it does not rely on user-defined parameters. The code and user guide can be downloaded at https://sourceforge.net/projects/snooper/ .
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 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 | 0.000 | 0.000 |
| 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".