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

• URL: http://arxiv.org/abs/2506.10271v3
• Date: Tue, 26 Aug 2025 04:00:29 GMT
• Title: Evaluating DNA function understanding in genomic language models using evolutionarily implausible sequences
• Authors: Shiyu Jiang, Xuyin Liu, Zitong Jerry Wang,
• Abstract summary: We introduce a benchmark called Nullsettes, which assesses how well models can predict in silico loss-of-function (LOF) mutations.<n>We find that most fail to consistently detect strong LOF mutations.<n>All models show a sharp drop in predictive accuracy as the likelihood assigned to the original (nonmutant) sequence decreases.
• Score: 0.25489046505746704
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• GRAPE: Heterogeneous Graph Representation Learning for Genetic Perturbation with Coding and Non-Coding Biotype [51.58774936662233]
  Building gene regulatory networks (GRN) is essential to understand and predict the effects of genetic perturbations.<n>In this work, we leverage pre-trained large language model and DNA sequence model to extract features from gene descriptions and DNA sequence data.<n>We introduce gene biotype information for the first time in genetic perturbation, simulating the distinct roles of genes with different biotypes in regulating cellular processes.
  arXiv  Detail & Related papers  (2025-05-06T03:35:24Z)
• UniGenX: Unified Generation of Sequence and Structure with Autoregressive Diffusion [61.690978792873196]
  Existing approaches rely on either autoregressive sequence models or diffusion models.<n>We propose UniGenX, a unified framework that combines autoregressive next-token prediction with conditional diffusion models.<n>We validate the effectiveness of UniGenX on material and small molecule generation tasks.
  arXiv  Detail & Related papers  (2025-03-09T16:43:07Z)
• Predicting gene essentiality and drug response from perturbation screens in preclinical cancer models with LEAP: Layered Ensemble of Autoencoders and Predictors [0.0]
  We introduce LEAP (Layered Ensemble of Autoencoders and Predictors), a novel ensemble framework to improve robustness and generalization.<n>LEAP consistently outperforms state-of-the-art approaches in predicting gene essentiality or drug responses in unseen cell lines, tissues and disease models.
  arXiv  Detail & Related papers  (2025-02-21T18:12:36Z)
• Learning to Discover Regulatory Elements for Gene Expression Prediction [59.470991831978516]
  Seq2Exp is a Sequence to Expression network designed to discover and extract regulatory elements that drive target gene expression.<n>Our approach captures the causal relationship between epigenomic signals, DNA sequences and their associated regulatory elements.
  arXiv  Detail & Related papers  (2025-02-19T03:25:49Z)
• GENERator: A Long-Context Generative Genomic Foundation Model [66.46537421135996]
  We present GENERator, a generative genomic foundation model featuring a context length of 98k base pairs (bp) and 1.2B parameters.<n>Trained on an expansive dataset comprising 386B bp of DNA, the GENERator demonstrates state-of-the-art performance across both established and newly proposed benchmarks.<n>It also shows significant promise in sequence optimization, particularly through the prompt-responsive generation of enhancer sequences with specific activity profiles.
  arXiv  Detail & Related papers  (2025-02-11T05:39:49Z)
• Generative Intervention Models for Causal Perturbation Modeling [80.72074987374141]
  In many applications, it is a priori unknown which mechanisms of a system are modified by an external perturbation.<n>We propose a generative intervention model (GIM) that learns to map these perturbation features to distributions over atomic interventions.
  arXiv  Detail & Related papers  (2024-11-21T10:37:57Z)
• Long-range gene expression prediction with token alignment of large language model [37.10820914895689]
  We introduce Genetic sequence Token Alignment (GTA), which aligns genetic sequence features with natural language tokens. GTA learns the regulatory grammar and allows us to further incorporate gene-specific human annotations as prompts. GTA represents a powerful and novel cross-modal approach to gene expression prediction by utilizing a pretrained language model.
  arXiv  Detail & Related papers  (2024-10-02T02:42:29Z)
• Semantically Rich Local Dataset Generation for Explainable AI in Genomics [0.716879432974126]
  Black box deep learning models trained on genomic sequences excel at predicting the outcomes of different gene regulatory mechanisms. We propose using Genetic Programming to generate datasets by evolving perturbations in sequences that contribute to their semantic diversity.
  arXiv  Detail & Related papers  (2024-07-03T10:31:30Z)
• VQDNA: Unleashing the Power of Vector Quantization for Multi-Species Genomic Sequence Modeling [60.91599380893732]
  VQDNA is a general-purpose framework that renovates genome tokenization from the perspective of genome vocabulary learning. By leveraging vector-quantized codebooks as learnable vocabulary, VQDNA can adaptively tokenize genomes into pattern-aware embeddings.
  arXiv  Detail & Related papers  (2024-05-13T20:15:03Z)
• A Poisson-Gamma Dynamic Factor Model with Time-Varying Transition Dynamics [51.147876395589925]
  A non-stationary PGDS is proposed to allow the underlying transition matrices to evolve over time. A fully-conjugate and efficient Gibbs sampler is developed to perform posterior simulation. Experiments show that, in comparison with related models, the proposed non-stationary PGDS achieves improved predictive performance.
  arXiv  Detail & Related papers  (2024-02-26T04:39:01Z)
• Efficient and Scalable Fine-Tune of Language Models for Genome Understanding [49.606093223945734]
  We present textscLingo: textscLanguage prefix ftextscIne-tuning for textscGentextscOmes. Unlike DNA foundation models, textscLingo strategically leverages natural language foundation models' contextual cues. textscLingo further accommodates numerous downstream fine-tune tasks by an adaptive rank sampling method.
  arXiv  Detail & Related papers  (2024-02-12T21:40:45Z)
• Predicting loss-of-function impact of genetic mutations: a machine learning approach [0.0]
  This paper aims to train machine learning models on the attributes of a genetic mutation to predict LoFtool scores. These attributes included, but were not limited to, the position of a mutation on a chromosome, changes in amino acids, and changes in codons caused by the mutation. Models were evaluated using five-fold cross-validated averages of r-squared, mean squared error, root mean squared error, mean absolute error, and explained variance.
  arXiv  Detail & Related papers  (2024-01-26T19:27:38Z)
• HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution [76.97231739317259]
  We present HyenaDNA, a genomic foundation model pretrained on the human reference genome with context lengths of up to 1 million tokens at the single nucleotide-level. On fine-tuned benchmarks from the Nucleotide Transformer, HyenaDNA reaches state-of-the-art (SotA) on 12 of 18 datasets using a model with orders of magnitude less parameters and pretraining data.
  arXiv  Detail & Related papers  (2023-06-27T20:46:34Z)
• Unsupervised language models for disease variant prediction [3.6942566104432886]
  We find that a single protein LM trained on broad sequence datasets can score pathogenicity for any gene variant zero-shot. We show that it achieves scoring performance comparable to the state of the art when evaluated on clinically labeled variants of disease-related genes.
  arXiv  Detail & Related papers  (2022-12-07T22:28:13Z)
• Generative Capacity of Probabilistic Protein Sequence Models [0.0]
  Potts models and variational autoencoders (VAEs) have recently gained popularity as generative protein sequence models (GPSMs) It is currently unclear whether GPSMs can faithfully reproduce the complex multi-residue mutation patterns observed in natural sequences arising due to epistasis. We develop a set of sequence statistics to assess the "generative capacity" of three GPSMs of recent interest.
  arXiv  Detail & Related papers  (2020-12-03T21:59:24Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.