# Evaluating DNA function understanding in genomic language models using evolutionarily implausible sequences

arXiv: https://arxiv.org/abs/2506.10271 · HTML: https://mlsanity.com/paper/2506.10271/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 40 references that are in the map.
- First posted on arXiv 12 Jun 2025.

## Builds on (most cited first, 10 of 10)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [BERT has a Mouth, and It Must Speak: BERT as a Markov Random Field Language Model](https://mlsanity.com/paper/1902.04094/) (2019)
- [DNABERT-2: Efficient Foundation Model and Benchmark For Multi-Species Genome](https://mlsanity.com/paper/2306.15006/) (2023)
- [Caduceus: Bi-Directional Equivariant Long-Range DNA Sequence Modeling](https://mlsanity.com/paper/2403.03234/) (2024)
- [GENERator: A Long-Context Generative Genomic Foundation Model](https://mlsanity.com/paper/2502.07272/) (2025)
- [A 5' UTR Language Model for Decoding Untranslated Regions of mRNA and Function Predictions](https://mlsanity.com/paper/2310.03281/) (2023)
- [Systems and Algorithms for Convolutional Multi-Hybrid Language Models at Scale](https://mlsanity.com/paper/2503.01868/) (2025)
- [METAGENE-1: Metagenomic Foundation Model for Pandemic Monitoring](https://mlsanity.com/paper/2501.02045/) (2025)

## Abstract (arXiv)

Genomic language models (gLMs) hold promise for generating novel, functional DNA sequences for synthetic biology. However, realizing this potential requires models to go beyond evolutionary plausibility and understand how DNA sequence encodes gene expression and regulation. We introduce a benchmark called Nullsettes, which assesses how well models can predict in silico loss-of-function (LOF) mutations, in synthetic expression cassettes with little evolutionary precedent. Testing 12 state-of-the-art gLMs, we find that most fail to consistently detect these strong LOF mutations. All models show a sharp drop in predictive accuracy as the likelihood assigned to the original (nonmutant) sequence decreases, suggesting that gLMs rely heavily on pattern-matching to their evolutionary prior rather than on any mechanistic understanding of gene expression. Our findings highlight fundamental limitations in how gLMs generalize to engineered, non-natural sequences, and underscore the need for benchmarks and modeling strategies that prioritize functional understanding.