Towards better prediction of Mycobacterium tuberculosis lineages from MIRU-VNTR data
Bibliographic record
Abstract
The determination of lineages from strain-based molecular genotyping information is an important problem in tuberculosis. Mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing is a commonly used molecular genotyping approach that uses counts of the number of times pre-specified loci repeat in a strain. There are three main approaches for determining lineage based on MIRU-VNTR data - one based on a direct comparison to the strains in a curated database, and two others, on machine learning algorithms trained on a large collection of labeled data. All existing methods have limitations. The direct approach imposes an arbitrary threshold on how much a database strain can differ from a given one to be informative. On the other hand, the machine learning-based approaches require a substantial amount of labeled data. Notably, all three methods exhibit suboptimal classification accuracy without additional data. We explore several computational approaches to address these limitations. First, we show that eliminating the arbitrary threshold improves the performance of the direct approach. Second, we introduce RuleTB, an alternative direct method that proposes a concise set of rules for determining lineages. Lastly, we propose StackTB, a machine learning approach that requires only a fraction of the training data to outperform the accuracy of both existing machine learning methods. Our approaches demonstrate superior performance on a training dataset collected in New York City over 10 years, and the improvement in performance translates to a held-out testing set. We conclude that our methods provide opportunities for improving the determination of pathogenic lineages based on MIRU-VNTR data.
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 | 0.004 | 0.012 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.004 | 0.002 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.001 | 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".