Genome-Factory is the first integrated Python library designed to streamline the entire workflow of genomic foundation model development, covering data collection, model tuning, inference, benchmarking, and interpretability. It provides automated pipelines for downloading and preprocessing genomic sequences, supports both full and parameter‑efficient fine‑tuning for diverse models, and offers tools for embedding extraction, DNA sequence generation, and flexible benchmarking. Additionally, the library includes an open‑source biological interpreter based on a sparse auto‑encoder, validated across compatibility, performance, and biological interpretation dimensions.
arXiv:2602. 17162v3 Announce Type: replace Abstract: Genomic Foundation Models (GFMs) typically rely on Masked Language Modeling (MLM) or Next-Token Prediction (NTP) to learn the "Laws of Nature".
VenusGenome is a gene‑resolution masked autoencoder that learns microbial genome representations by combining visible‑only encoding with genome‑aware attention, which incorporates contig boundaries, relative strand orientations, and gene positions. The model is pretrained on over 1.7 million microbial genome assemblies using frozen protein embeddings as inputs, and its frozen representations outperform other evaluated models on nine of eleven primary metrics for both genome‑level phenotype prediction and gene‑level tasks. Ablation studies confirm the importance of genome‑aware attention and demonstrate that higher masking ratios and masked autoencoding yield better performance than BERT‑style masked modeling.
arXiv:2609.14882v1 Announce Type: cross Abstract: Nucleotide sequence analysis is central to problems spanning regulatory genomics, evolutionary biology, and phenotype prediction. Classical bioinform...
arXiv:2608.07632v2 Announce Type: replace-cross Abstract: Image-based profiling captures rich phenotypic signatures for drug discovery and functional genomics. Large public datasets like JUMP Cell Pa...
arXiv:2606. 07760v1 Announce Type: new Abstract: Understanding cellular phenotypes and how they respond to perturbations is critical for disease biology and therapeutic design.
arXiv:2606. 05139v1 Announce Type: new Abstract: The rapid advancement of high-throughput sequencing has led to large, high-dimensional omics datasets.
RAGenome is a retrieval‑based genomic language model that extends pretraining to contexts 100 times longer than existing MSA‑based models, enabling it to capture both cross‑species evolutionary relationships and within‑species long‑range interactions. Trained on whole‑genome alignments from 100 vertebrates, it improves gene‑finding performance from 0.45 to 0.60 while staying competitive on evolutionary tasks such as prioritizing pathogenic variants. The model offers competitive performance at a fraction of the training cost, unifying evolutionary modeling and long‑range capabilities in a single scalable framework.
CellMSA introduces a novel single‑cell representation learning framework that leverages a multiple‑sequence‑alignment‑inspired context model. For each target cell, it retrieves relevant cells across batches and related cell types, summarizing cross‑cell patterns into a context‑dependent gene‑pair representation that is fed into a pair‑aware encoder. Pretraining on a massive human single‑cell corpus (≈109 million cells) and subsequent benchmarks demonstrate consistent performance gains over existing methods.
arXiv:2607. 08803v1 Announce Type: cross Abstract: The push toward large language models for biology (BioLM) has created a need for training corpora that can endow models with a genuine understanding of biology.
GENSCRIPT is an inference‑only pipeline that generates synthetic data without training a generative model. It creates a deterministic statistical profile of the source data, uses a language model to infer field semantics and cross‑column constraints, and then compiles these into an executable sampler that works for single‑table, temporal, and relational data. The method builds generators in minutes, samples large datasets quickly, and achieves fidelity comparable to leading methods while preserving key data relationships such as 1‑to‑1 mappings and primary‑foreign key constraints.
arXiv:2606. 16540v1 Announce Type: cross Abstract: Biomolecular sequence models are increasingly reused outside the studies in which they were introduced, but public checkpoints rarely preserve the execution context needed to inspect source-defined behavior, adapt models to new assays, compare models under shared task definitions or deploy biological predictions.
SoftGene is a new framework that enhances gene set annotation by combining protein language model representations with a hybrid prompting scheme. It uses a hierarchical attention-based encoder built on ESM to encode gene sets from protein amino acid sequences, then merges soft prompts derived from these embeddings with hard prompts generated by a large language model. The approach is evaluated on Gene Ontology and MSigDB datasets, showing that integrating protein-sequence information with textual context improves overall annotation performance, though the benefit varies across biological domains.