VESTIGE: A Knowledge-Guided Masking Strategy for Corruption-Aware Fine-Tuning of Genomic Transformers, Validated on Ancient DNA Reconstruction

Abstract:Standard masked-language-model fine-tuning applies a uniform masking probability across every token position, assuming reconstruction difficulty is position-agnostic. When the degradation process is characterised and concentrated at predictable positions, this assumption fails: at peak damage sites the model can underperform a frequency-matched random predictor. We introduce VESTIGE, a parameter-free, drop-in replacement for the standard MLM collator that aligns the masking distribution with an empirically measured per-position corruption profile. We apply it to ancient DNA (aDNA) reconstruction, where cytosine deamination produces a position-dependent C-to-T / G-to-A gradient quantified per-position by mapDamage2. Rescaling so the mean C/G masking rate equals 15% - identical to standard MLM - isolates spatial redistribution as the sole variable, with model, data, seed, and hyperparameters held fixed across both DNABERT-2 runs on a mammoth CDS corpus (two specimens, seven genes). Across six terminal-zone widths and 626 paired windows, VESTIGE leads standard MLM at every width (Delta = +4.18 to +10.35 pp, all p < 10^-8), cuts validation cross-entropy by 13% (3.274 vs. 3.757), and yields ESMFold reconstructions with TM-score > 0.95 across all six reconstructions (three genes) even under damage amplified 10-30x beyond authentic PMD rates. A 1D CNN biosecurity classifier returns AUC = 0.935 and clears 98.2% of reconstructed windows, the 1.76% remainder attributable to reference-genome features, not reconstruction artefacts. The principle is domain-agnostic: any measurable position- or context-specific corruption profile - FFPE, bisulfite, metagenomic, or nanopore - substitutes directly for the PMD array, making VESTIGE a knowledge-guided training routine for intelligent systems operating on degraded or noisy sequence inputs.

Machine Learning (cs.LG)

arXiv:2607.27712 [cs.LG]

(or arXiv:2607.27712v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2607.27712 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.