Deciphering how DNA sequence encodes gene regulation remains a central challenge in biology. Advances in machine learning and functional genomics have enabled sequence-to-function (seq2func) models that predict molecular regulatory readouts directly from DNA sequence. These models are now widely used for variant effect prediction, mechanistic interpretation, and regulatory sequence design. Despite strong performance on held-out genomic regions, their ability to generalize across genetic variation and cellular contexts remains inconsistent. Here we examine how architectural choices, training data, and prediction tasks shape the behavior of seq2func models. We synthesize how interpretability methods and evaluation practices have probed learned cis-regulatory organization and highlighted systematic failure modes, clarifying why strong predictive accuracy can fail to translate into robust regulatory understanding. We argue that progress will require reframing seq2func models as continually refined systems, in which targeted perturbation experiments, systematic evaluation, and iterative model updates are tightly coupled through AI-experiment feedback loops. Under this framework, seq2func models become self-improving tools that progressively deepen their mechanistic grounding and more reliably support biological discovery.
Phenotypic divergence of Homo sapiens is driven by the evolution of human-specific genomic regulatory networks via two mechanistically distinct pathways of creation of divergent regulatory DNA sequences
Analysis of evolutionary origins of genomic loci harboring 59,732 candidate human-specific regulatory sequences identifies genetic divergence patterns during evolution of Great Apes
Functional architecture and global properties of the Corynebacterium glutamicum regulatory network: novel insights from a dataset with a high genomic coverage
Genomics-guided drawing of malignant regulatory signatures revealed a pivotal role of human stem cell-associated retroviral sequences (SCARS) and functionally-active hESC enhancers
Institute of Engineering in Medicine University of California, San Diego（加州大学圣地亚哥分校工程医学院）；OncoSCAR, Inc.（OncoSCAR公司）
Highly conserved sequence-specific double-stranded DNA binding networks contributing to divergent genomic evolution of human and chimpanzee brain development
Institute of Engineering in Medicine University of California, San Diego（加州大学圣地亚哥分校工程医学院）
Embedding Is (Almost) All You Need: Retrieval-Augmented Inference for Generalizable Genomic Prediction Tasks
Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology（孟加拉工程与技术大学计算机科学与工程系）；Department of Computer Science and Engineering, BRAC University（布拉克大学计算机科学与工程系）
DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA
Department of Computer Science, School of Engineering, Stanford University（计算机科学系，工程学院，斯坦福大学）；Department of Genetics, School of Medicine, Stanford University（遗传学系，医学院，斯坦福大学）；Department of Pathology, School of Medicine, Stanford University（病理学系，医学院，斯坦福大学）
Combining detection and reconstruction of correlational and quasi-periodic motifs in viral genomic sequences with transitional genome mapping: Application to COVID-19
基因组自身的过渡自同构映射（Transitional Automorphic Mapping of the Genome on Itself，TAMGI）方法旨在结合病毒基因组RNA/DNA序列中相关和准周期模体的检测和重建。通过TAMGI重建的基序对于插入缺失和点突变是稳健的，并且可以尝试作为推定的治疗靶标。我们开发并测试了相关的理论和统计标准的TAMGI应用程序。TAMGI的应用通过研究包装在类似病毒的螺旋衣壳中的严重急性呼吸道综合征冠状病毒SARS-CoV和SARS-CoV-2（后者的冠状病毒SARS-CoV-2负责COVID-19大流行）的基因组中的基序来说明。这样的核糖核衣壳被转运到球形膜被膜与掺入刺突糖蛋白。另外两个例子涉及二十面体衣壳病毒的基因组，卫星烟草花叶病毒（STMV）和噬菌体PHIX 174。这些病毒基因组中的部分准周期基序是由于基因组ssRNA/ssDNA与核衣壳蛋白之间的弱特异性合作相互作用而进化的。衣壳的对称性导致相关基因组序列中特定准周期基序的自然选择。一般来说，TAMGI为研究许多分子机制提供了一个方便的工具，这些机制涉及准周期基序和完整重复序列、基因组组织、顺式/反式调节元件的背景分析、数据挖掘以及基因组序列中的相关性。
Semantically Rich Local Dataset Generation for Explainable AI in Genomics
LASIGE, Faculdade de Ciências da Universidade de Lisboa（里斯本大学理学院LASIGE实验室）；Instituto de Medicina Molecular João Lobo Antunes, Faculdade de Medicina da Universidade de Lisboa（里斯本大学医学院若昂·洛博·安图内斯分子医学研究所）；Mondego Science（蒙德戈科学公司）
STRAND: Sequence-Conditioned Transport for Single-Cell Perturbations
Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA；Merck \& Co., Inc., Cambridge, MA, USA；Department of Biomedical Engineering, Johns Hopkins University, Baltimore, MD, USA；Kempner Institute for the Study of Natural；Artificial Intelligence, Harvard University, Allston, MA, USA；Broad Institute of MIT
GENERator: A Long-Context Generative Genomic Foundation Model
Equal Contribution 12pt Senior Authorship 0.15in 1Alibaba Cloud Computing, Beijing, China 0.05in 2Zhongguancun Academy, Beijing, China 0.05in 3Zhongguancun Institute of Artificial Intelligence, Beijing, China 0.05in 4University of Science；Technology of China, Hefei, China 0.05in 5State Key Laboratory of Genome；Multi-omics Technologies, Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Key Laboratory of Livestock；Poultry Multi-Omics of MARA, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen, China 0.05in 6Innovation Group of Pig Genome Design；Breeding, Research Centre for Animal Genome, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen, China 0.05in 7The Hong Kong University of Science；Technology (Guangzhou), Guangzhou, China 0.05in 8The Hong Kong University of Science；Technology , Hong Kong SAR, China 0.05in 9Mila - Qu\'ebec AI Institute, Montr\'eal, Canada 0.05in 10University of Montr\'eal, Montr\'eal, Canada 0.05in 11HEC Montr\'eal, Montr\'eal, Canada 0.05in 12CIFAR AI Chair, Canada 0.05in 13Carnegie Mellon University, Pittsburgh, USA 0.15in to
