Peer Revenue Publish or profit

Papers

Fresh from arXiv, typeset properly. Highlighted words trade on the market.

Quantitative Biology Genomics

q-bio.GN Watch Genomics

Sunday 27 September

Genomics · Cross-listed from cs.CR

CipherGenome: Homomorphic Inference for Genomic Mixture-of-Experts

Guang Yang and Fengchen Liu

Genome foundation models are growing into sparse mixture-of-experts (MoE) networks whose expert weights no longer fit on the machines that hold the sequences, yet sending a private genome to rented accelerators exposes it: we show that a single server hosting one expert recovers the input nucleotides with 99.8% top-1 accuracy. We present CipherGenome, a protocol that keeps the embedding, attention and router of a 15.1B-parameter MoE genome model on a trusted thin client and outsources every expert projection, 95.8% of the parameters, to untrusted and possibly colluding GPU servers under module-LWE encryption. The design exploits three structural facts: expert layers are linear between two SwiGLU gates, expert weights are public, and GPU integer tensor cores can evaluate a ciphertext-weight product exactly modulo $2^{48}$ in a single GEMM. The client evaluates the nonlinearity exactly and re-encrypts with fresh secrets, so no polynomial approximation or bootstrapping is ever needed. On 72 windows from 12 bacterial genomes, encryption adds $2.54 \times 10^{-4}$ nats per token of KL divergence (95% CI upper bound $3.95 \times 10^{-4}$), below a pre-registered non-inferiority margin and indistinguishable from bf16 inference, while the same inversion attack falls to chance level. A reusable public hint cuts end-to-end latency by 3.54 times, wire compression reduces traffic 6.8 times, per-layer padding reduces routing leakage from 54.9% to 8.9% accuracy, and HE-compatible int4 experts remain non-inferior to their plaintext counterparts. Per expert and token, the server-side cost is more than six orders of magnitude below a CKKS baseline.
Read the full abstract

Genomics · Cross-listed from cs.CR

GenomeOcean Anywhere: Private WebGPU Inference for Genome MoEs

Guang Yang and Fengchen Liu

Genome foundation models are most useful where sequences are generated, yet the largest models need datacenter accelerators and a place to send private DNA. We ask whether a 15-billion-parameter genome mixture-of-experts (MoE) model can instead run on volunteers' web browsers, with the experts spread across many untrusted devices, without changing its predictions and without revealing the sequence to any single device. We build a system in which a trusted coordinator runs attention and routing while browser workers run every expert feed-forward network through hand-written WebGPU kernels, and we protect the expert inputs with real-valued Lagrange coded computing: each worker receives only a Gaussian-padded share, computes the expert's linear maps, and the coordinator decodes from any two of three workers. On GenomeOcean-MoE (8 experts, top-2 routing, 24 layers), the browser path matches native llama.cpp at every quantization level, the distributed path stays at the BF16 numerical noise floor (KL 0.0036 nats per token), and an unfitted latency model predicts decode time within 0.74% (median) under emulated wide-area links. We first show that plaintext expert inputs are not private: a probe recovers the token from a single vector at every depth, and one worker can identify the source genome from 300 unordered tokens with 92% accuracy. With coded experts, an adaptive attacker trained on shares falls to the most-frequent-token baseline, one worker's information about each token is bounded below one bit per forward pass, and the fidelity cost stays below the BF16 noise floor; in Chrome, coded decoding runs at 220 to 376 ms per token, depending on how much of the routing is hidden, and continues without replicas when a worker fails.
Read the full abstract

Genomics · Cross-listed from cs.CR

GenoTrace: Inheritable Watermarks for Genome Foundation Model Distillation

Guang Yang and Fengchen Liu

Can a genome model retain a detectable record of the synthetic sequences used to train it? We study watermark inheritance through distillation with GenoTrace, a codon-aware extension of green-list watermarking. Two token-level factors modulate the teacher's generation bias using codon position and organism-specific codon usage. The resulting sequences train a smaller student, whose outputs are audited without an active watermark processor. In a three-seed GenomeOcean-500M-to-100M experiment, the joint configuration achieves a mean audit score of 17.88 and 94.5% detection at a fixed threshold. It retains 49.0% detection after key-aware token substitution, compared with 0% for the available single-seed plain-watermark comparator, and 47.0% after combined mechanism-targeted nucleotide edits. Additional experiments establish inherited signal across five organism-conditioned datasets and teacher-student size ratios up to 40. Component ablations and computational sequence-quality assays reveal distinct operating points for detection strength and coding coverage. GenoTrace provides a practical token-level construction and an empirical account of how genomic structure shapes inherited watermark signals. The findings concern shared-tokenizer distillation and the tested editing procedures, with calibration and biological utility treated as separate evaluation requirements.
Read the full abstract

Saturday 26 September

Genomics

BarcodeMAE+: Rethinking Masked Pretraining and Global Representations for DNA Barcode Foundation Models

Monireh Safari, Pablo Millan Arias, Scott C. Lowe et al. (+3 more)

Many DNA foundation models are pretrained by masking parts of a sequence and asking the model to reconstruct them. Standard masked pretraining exposes the encoder to special [MASK] tokens that are absent at inference, creating a mismatch between training and downstream use. The role of an explicit global sequence representation such as a [CLS] token and how it should be trained also remain poorly understood for DNA barcodes. We introduce BarcodeMAE+ and study model architecture, global [CLS] representation, and auxiliary pretraining objectives across arthropod COI (BIOSCAN-5M) and fungal ITS (UNITE+INSD) barcodes. Across both barcode regions, the encoder-decoder MAE-LM architecture outperforms its matched encoder-only counterpart in nearly all evaluated configurations, supporting MAE-LM as an effective architectural design for DNA barcode foundation models. A trained global [CLS] representation provides substantial additional gains: on BIOSCAN-5M, [CLS] accuracy increases from 47.53% without an auxiliary objective to 80.65% with cross-entropy genus classification. The best auxiliary objective is region-dependent: cross-entropy performs best on BIOSCAN-5M, whereas pairwise same-genus classification performs best on UNITE+INSD, reaching 73.19% on Yeast and 63.07% on Filamentous Fungi. BarcodeMAE+ outperforms published DNA foundation model baselines on BIOSCAN-5M and achieves the highest Yeast accuracy among the evaluated UNITE+INSD baselines using frozen encoder representations. Similarity-weighted softmax KNN voting further stabilizes accuracy as neighbourhood size increases. Overall, encoder-decoder masked pretraining and an explicitly trained global representation are strong design choices for DNA barcode foundation models, while the optimal objective for learning that representation depends on the biological domain.
Read the full abstract

All authors Monireh Safari, Pablo Millan Arias, Scott C. Lowe, Lila Kari, Angel X. Chang, Graham W. Taylor

Genomics · Cross-listed from cs.AI

AmbiModBench: Benchmarking Gene Perturbation Prediction Beyond Shared Responses

Sikai Huang, Zhiwen Yang, Kai Yu et al. (+2 more)

Predicting cellular responses to genetic perturbations helps prioritize experiments in single-cell genomics, where exhaustive measurement is infeasible. While computational models increasingly predict these responses, three evaluation deficiencies obscure what their scores demonstrate. First, absolute metrics cannot separate target-specific predictions from a shared background response. Second, common metrics remain high under gene shuffling, so gene-level accuracy is never verified. Third, a score at one training size says nothing about coverage, which depends on representation-space proximity and response-constraining power. We propose AmbiModBench, a specificity-aware, gene-resolved and coverage-aware benchmark. It pairs every score with a training-mean reference fitted on the same split, screens each readout by gene-coordinate permutation, and links embedding distance to response variation. Across K562, RPE1 and Norman, strong absolute scores largely reflect shared background rather than target-specific learning. Widely used readouts track response magnitude distributions rather than the affected genes. Detectable gain follows representation-space coverage rather than training-set size. Nonetheless, on RPE1 the protocol yields a reproducible target-specific gain across five additional splits and three gene selections, which absolute scores alone cannot distinguish from shared background.
Read the full abstract

All authors Sikai Huang, Zhiwen Yang, Kai Yu, Jiayuan Chen, Stan Z. Li

Friday 25 September

Genomics

CANN: Commutative Algebra Neural Networks for Virus Classification

Mushal Zia, Faisal Suwayyid and Guo-Wei Wei

The extensive diversity of viral genomes challenges alignment-free representations to capture both nucleotide composition and the higher-order organization of recurring sequence patterns. Conventional $k$-mer representations quantify pattern abundance but do not explicitly encode how these patterns are organized across scales. Here, we introduce CANN, an alignment-free framework that combines persistent commutative-algebraic descriptors with neural representation learning for viral family classification. CANN uses persistent facet descriptors to encode the multiscale organization of $k$-mer patterns and complementary $k$-mer counts to capture sequence composition. Convolutional and Transformer encoders independently learn latent representations of these features, and cosine distances within their respective embedding spaces are averaged before $k$-nearest-neighbor classification. Across multiple NCBI viral genome collections, CANN achieves robust family-level performance, with the fusion of encoder-specific distances consistently improving classification relative to either encoder alone. Evaluation on genomes present in the 2026 NCBI collection but absent from the 2024 reference set further supports generalization to unseen members of families already represented in the reference data. Together, these findings show that persistent commutative-algebraic descriptors can complement neural representations of viral genomes, linking algebraic descriptions of genomic organization with data-driven classification.
Read the full abstract

Thursday 24 September

Genomics

EMMA: an R/Bioconductor package to automate tracking of metadata in functional enrichment analyses

Najla Abassi, Annekathrin Silvia Nedwed and Federico Marini

Summary: Functional enrichment analysis (FEA) is a widely used approach for interpreting high-throughput omics data. However, essential methodological details, such as software versions, analysis parameters, and annotation database releases among others, are often incompletely reported, limiting the reproducibility and transparency of enrichment analyses and complicating the assessment of potentially problematic methodological choices. Here we present EMMA, an R/Bioconductor package that integrates with existing FEA tools and automatically captures provenance metadata, such as annotation metadata, software version, and parameters, during the analysis runtime. Our package provides utilities for accessing and exporting the recorded metadata to facilitate transparent reporting and preserve provenance required for reproducible enrichment analyses. This also enables auditing of the results while remaining compatible with existing Bioconductor workflows. Availability and implementation: EMMA is available on Bioconductor under the MIT license (https: //bioconductor.org/packages/EMMA), with its development version also available on GitHub (https: //github.com/imbeimainz/EMMA).
Read the full abstract

Wednesday 23 September

Genomics · Cross-listed from cs.AI

BaseCamp — An Agentic AI Framework for Automating DNA Sequencing Data Pipelines

Eranga Bandara, Xueping Liang, Asanga Gunaratna et al. (+19 more)

DNA sequencing pipelines, spanning quality control, alignment, variant calling, and annotation, are now reliably executed by workflow management systems that orchestrate established bioinformatics tools at scale. What remains manual is the decision layer surrounding that execution: selecting quality thresholds appropriate to a sample and platform, adjudicating borderline variant calls, diagnosing anomalies, and determining which findings warrant expert review. These decisions are repetitive, judgment-intensive, inconsistent across operators, and frequently undocumented. This paper introduces BaseCamp, a novel agentic AI framework for automating the decision layer of DNA sequencing pipelines. The framework decomposes the pipeline into six specialized AI agents, covering sample intake and quality control, alignment, variant calling, annotation, cross-stage monitoring, and reporting. Critically, BaseCamp agents do not perform sequence analysis: established tools execute alignment, calling, and annotation, while the agents select among them, configure them, interpret their output, and decide what follows. This confines language model reasoning to the judgment layer where it is reliable and preserves the reproducibility existing tooling guarantees. Agent reasoning is powered by a consortium of fine-tuned, domain-specialized large language models coordinated by a central reasoning LLM, executing locally so no sequencing data leaves the operating environment, under human-in-the-loop orchestration. Evaluation shows agent-generated configurations are concordant with expert practice, that an explicit filtering ledger renders inspectable what filtering otherwise removes without trace, and that cross-stage anomaly detection surfaces conditions execution monitoring misses. BaseCamp offers a generalizable blueprint for agentic automation of scientific data pipelines.
Read the full abstract

All authors Eranga Bandara, Xueping Liang, Asanga Gunaratna, Tharaka Hewa, Abdul Rahman, Peter Foytik, Safdar H. Bouk, Sachini Rajapakse, Isurunima Kularathna, Pramoda Karunarathna, Chalani Rajapakse, Ng Wee Keong, Kasun De Zoysa, Amin Hass, Wathsala Herath, Ross Gore, Ravi Mukkamala, Nihal Siriwardanagea, Gihan Siriwardanagea, Aruna Withanage, Nilaan Loganathan, Sachin Shetty

Genomics

Motif-Vocab: StatisticallyCalibrated Transcription-Factor-Identity Tokenization forGenomic Language Models

Liangyu Li and Michael White

Tokenization is a central design choice in genomic language models, yet most deoxyribonucleic acid (DNA) tokenizers use characters, fixed-length k-mers, or frequency-derived subwords without explicitly using prior information about the specificity of DNA-binding regulatory factors. We introduce Motif-Vocab, a biologically informed tokenizer that scans both DNA strands for statistically calibrated motif matches, emits transcription-factor (TF) identity tokens, and applies nucleotide, $k$-mer, or byte-pair encoding (BPE) to unmatched sequence. Motif-specific null distributions put position-weight matrices (PWMs) of different lengths and degeneracy on a common significance scale; deterministic overlap rules make the representation reproducible. In controlled Bidirectional Encoder Representations from Transformers (BERT) pretraining on two billion base pairs, real motif libraries outperform randomized-motif controls on 54 of 55 in-scope downstream tasks. On a motif-disjoint recognition task derived from DART-Eval Task 2, TF-specific tokens improve macro-F1 by 0.040 over a position-matched generic motif token and by 0.033 over a matched no-motif tokenizer (95% bootstrap confidence interval: 0.027–0.038). Motif tokens also receive stronger attribution and produce larger occlusion effects than shuffled controls. Dense no-motif tokenizers remain strong general-purpose baselines, including a near-tie on the five-task BERT-base panel. Thus, Motif-Vocab is not a universal accuracy replacement; it is a targeted, interpretable inductive bias for motif-sensitive genomic modeling.
Read the full abstract

Genomics · Cross-listed from q-bio.QM

BreCol: Benchmarking Classical and Deep-Learning Methods for Microbiome-Based Cancer Detection

Jeffrey M. Dick

DNA sequencing of the gut microbial community shows promise for cancer detection, but questions remain about the generalizability of results across studies. We propose BreCol, a benchmark of 2,040 16S rRNA gene sequencing runs across 26 studies spanning breast cancer, colorectal cancer, and healthy cohorts. Train-test splits are made within pre-2023 studies, while holdout evaluation uses studies from 2023 onward, reflecting temporal separation from training data. Classical models reach test/holdout AUCs of 0.77/0.60 for cancer diagnosis and 1.00/0.83 for cancer type prediction. We train the models on both cancer types simultaneously and find that colorectal cancer is often easier to detect than breast cancer. We also evaluate two deep learning models: HyenaDNA, a long-range sequence model that pools hidden states for classification, and SetBERT, a transformer that produces contextualized embeddings over sets of reads. Both deep learning models underperform the best classical methods on holdout data, though tuning training set size and the classification head yields modest gains. Our classical pipeline uses unsupervised clustering to derive features from tetramer frequencies, preserving within-run compositional signal and achieving near state-of-the-art performance without relying on taxonomic assignments. BreCol data and associated code are publicly available.
Read the full abstract

Thursday 17 September

Genomics · Cross-listed from quant-ph

Hybrid quantum-classical attention for histopathology-based molecular profiling in data-limited cancers

Kahn Rhrissorrakrai, Aritra Bose, Aldo Guzman-Saenz et al. (+2 more)

Molecular profiling from routine histopathology could expand access to precision oncology when sequencing is unavailable, tissue is limited, or training cohorts are small. We developed a hybrid quantum-classical strategy that replaces softmax attention in a transformer for histopathology-based gene expression prediction with a quantum-derived doubly stochastic matrix (QDSM). Across 29 cancer cohorts from The Cancer Genome Atlas and an independent pancreatic cancer cohort from the Clinical Proteomic Tumor Analysis Consortium, QDSM attention produced selective gains, with the largest relative improvements in smaller, data-limited cohorts, including adrenocortical carcinoma and uveal melanoma. Rather than improving transcriptome-wide performance uniformly, QDSM redistributed predictive accuracy across genes and pathways, improving biologically relevant targets in some tumor contexts while worsening others. In adrenocortical carcinoma, preferentially improved genes were enriched for adverse overall-survival associations, linking enhanced molecular inference to prognostically relevant biology. In pancreatic cancer transfer experiments, QDSM improved selected metabolic and lineage-associated genes but did not consistently improve performance under cross-cohort shift. Leave-one-cancer-out mixed-effects analysis showed that baseline molecular features predicted part of the gene-level benefit, while residuals identified cancer-specific programs that improved more or less than expected. Separate experiments on IBM quantum processors recovered the doubly stochastic matrix primitive underlying the attention mechanism. These findings position QDSM attention as a context- and target-dependent strategy for image-based molecular profiling and molecular triage when direct testing is unavailable, incomplete, or impractical.
Read the full abstract

All authors Kahn Rhrissorrakrai, Aritra Bose, Aldo Guzman-Saenz, Filippo Utro, Laxmi Parida

Genomics

Large Language Model Agents for Evidence Based Genetic Disease Severity Classification

Tohid Ghasemnejad, Ahmadreza Argha, Mark Grosser et al. (+7 more)

Disease severity classification for genetic conditions is subjective and labor-intensive, creating bottlenecks in genomic screening, where commercial panels vary widely in size and overlap. We developed an autonomous AI agent integrating Reasoning and Acting (ReAct) with Retrieval-Augmented Generation (RAG) to classify 10,211 Human Phenotype Ontology terms. It uses American College of Medical Genetics (ACMG)-endorsed severity guidelines and American College of Obstetricians and Gynecologists (ACOG) quality-of-life criteria to retrieve PubMed literature, generate interpretable reasoning chains, and independently verify claims. At the phenotype level, using expert-curated cohorts, the agent achieved 93.55% accuracy (MCC 0.9237) with 82.6% to 91.4% of claims supported by direct evidence or valid inferences. Gene-level severity was aggregated across 8,738 pairs, identifying 3,283 autosomal recessive pairs with severe or profound presentations. External validation showed 95.2% concordance with Mackenzie's Mission gene list. This system enables standardized panel design by providing reliable, automated classification supported by direct evidence.
Read the full abstract

All authors Tohid Ghasemnejad, Ahmadreza Argha, Mark Grosser, John Wang, Min Yang, Thantrira Porntaveetus, Tony Roscioli, Nigel H. Lovell, Mahmoud Aarabi, Hamid Alinejad-Rokny

Wednesday 16 September

Genomics

STUART: Sequence Triage and qUAntification of Read Transcripts for Rapid Ionizing Radiation Exposure Assessment

Tomasz Strzoda, Lourdes Cruz-Garcia, Mustafa Najim et al. (+2 more)

Rapid medical triage following ionizing radiation exposure is critical for emergency management, yet traditional alignment-based bioinformatics are too computationally intensive for mass-casualty scenarios. To address this, we developed STUART (Sequence Triage and qUAntification of Read Transcripts), a mapping-free machine learning framework optimized for mobile biological dosimetry. Inspired by Natural Language Processing (NLP), the system converts raw sequencing reads into k-mer-based numerical profiles, completely bypassing standard alignment. While the architecture is universally applicable to any transcriptomic biomarker, this study focused on radiation exposure using the FDXR gene model. Evaluating Logistic Regression, Random Forest, and XGBoost, advanced signature selection strategies drastically reduced the initial 1024-dimensional feature space by over 98%. Highly robust performance - characterized by near-perfect balanced accuracy and F1-scores within the 95-100% range - was consistently achieved while retaining as few as 17 transcriptomic signatures. Crucially, learning curve analysis demonstrated that complete signal stabilization requires aggregating merely 1000 potentially related reads. Furthermore, external validation on an independent dataset yielded over 99% specificity, confirming the tissue-agnostic nature of the extracted signatures despite different cellular origins. The framework's exceptionally low data threshold enables a real-time, analyze-as-you-sequence diagnostic paradigm compatible with portable sequencers. By minimizing time-to-decision, this decentralized tool bridges the gap between advanced biomarkers and practical on-site biomonitoring, offering a scalable foundation for rapid epidemiological response and routine occupational radiation monitoring.
Read the full abstract

All authors Tomasz Strzoda, Lourdes Cruz-Garcia, Mustafa Najim, Christophe Badie, Joanna Polanska

Genomics · Cross-listed from cs.AI

HPOQuest: A Rare-Disease Diagnostic Agent Using Active Phenotype Acquisition

Kamilia Zaripova, Nassir Navab, Azade Farshad et al. (+1 more)

More than 300 million people worldwide are affected by one of over 7,000 known rare diseases, yet diagnosis remains difficult because patients initially present with incomplete and heterogeneous phenotypes. We present HPOQuest, a training-free framework for sequential phenotype acquisition in rare-disease diagnosis. Starting from a small set of observed patient phenotypes, HPOQuest maintains a probabilistic disease ranking and iteratively selects informative follow-up questions to support clinicians during patient assessment. Confirmed phenotypes update the disease ranking, while all responses update the candidate question set. Across four benchmark cohorts, HPOQuest substantially improves diagnosis from sparse initial phenotypes, with gains of up to 30% points at Recall@1 and 45% points at Recall@5. These results demonstrate that sequential phenotype acquisition can substantially improve rare-disease diagnosis from limited initial clinical evidence.
Read the full abstract

All authors Kamilia Zaripova, Nassir Navab, Azade Farshad, Annalisa Marsico

Genomics

PlainMap: a lightweight, restartable mapping pipeline for ancient and modern DNA

Michael V. Westbury

Mapping sequencing reads to a reference genome requires preprocessing and alignment choices that can vary with library type, fragment length, sequencing platform, and reference genome. These considerations are particularly important for ancient and historical DNA, where short and damaged fragments can make the most appropriate mapping strategy difficult to determine a priori. We present PlainMap, a lightweight and restartable mapping pipeline for modern and degraded DNA sequencing data. PlainMap accepts a simple manifest of FASTQ files, automatically identifies single-end and paired-end data from read headers, supports mixed sequencing platforms, and provides alternative mapping strategies for modern and degraded DNA. Deterministic chunking and checkpoint-based execution allow large analyses to resume after interruption, while optional pilot subsampling enables empirical comparison of mapping strategies using identical subsets of raw fragments. PlainMap produces duplicate-filtered BAM files together with fragment-aware mapping and coverage statistics. Evaluation using heterogeneous sequencing data confirmed the expected behaviour of the three mapping modes, while adaptive chunking reduced peak memory use by approximately 25% and allowed an interrupted analysis to resume from completed mapping chunks. PlainMap is implemented as a single Bash script and is freely available at https://github.com/BiodiversityExtinction/PlainMap.
Read the full abstract

Tuesday 15 September

Genomics · Cross-listed from q-bio.QM

GIA: Germline-Informed Aging with AlphaGenome Finds Genetically Regulated CpGs

Sean Lim

Epigenetic clocks estimate age and aging-related phenotypes from DNA methylation at selected CpG sites, but the extent to which these inputs are influenced by germline genetic variation is unclear. Because methylation at many CpGs is genetically regulated, some between-person variation in clock estimates may reflect inherited genetic differences rather than aging-related change alone. Here we developed GIA (Germline-Informed Aging), a framework that maps CpGs selected from 13 published epigenetic clocks to blood methylation quantitative trait loci (meQTLs) and scores associated genetic variants with AlphaGenome. We show that clock CpGs were enriched for blood meQTLs relative to matched unused Illumina 450k probes (62.7% versus 39.6%; OR 2.57), across multiple clock families, suggesting that age-informative methylation sites are heavily influenced by germline genetic variation. Ranking by predicted chromatin effect isolated rs10190186, a cis-acting variant at FHL2 predicted to increase blood chromatin accessibility (ATAC +1.00; DNase +1.64) and FHL2 RNA (+0.30). This locus illustrates how inherited variation may shape methylation features repeatedly used by epigenetic clocks, motivating direct tests of whether such variants shift baseline clock estimates or longitudinal aging trajectories.
Read the full abstract

Genomics · Cross-listed from cs.LG

HyCoSeq: Contextual Hyperbolic Representation Learning for Genomic Sequences

Chenhao Zeng, Zhibin Pu and Shufei Ge

Hyperbolic geometry provides a natural inductive bias for genomic representation learning, but existing hyperbolic genomic models primarily use Lorentz convolutions to learn local sequence representations, while their residual pathways do not directly aggregate full Lorentz representations. We propose HyCoSeq, a contextual hyperbolic representation learning framework for genomic sequences. HyCoSeq incorporates weighted Lorentzian residual aggregation into multi-curvature Lorentz encoding, allowing full Lorentz representations to participate directly in geometry-consistent local aggregation. It further introduces a bidirectional long short-term memory network that integrates information from both sequence directions to learn contextual relationships among local representations at different positions within a genomic sequence, thereby extending local hyperbolic convolutional encoding to sequence-level contextualized representations. Extensive experiments across diverse genomic tasks show that HyCoSeq outperforms existing hyperbolic baselines and, without large-scale genomic pretraining, achieves competitive performance against substantially larger pretrained DNA language models.
Read the full abstract

Monday 14 September

Genomics

Democratizing Clinical Tumor Whole Genome Sequencing: 18-hour End-to-end Analysis via Trillion-parameter Large Language Models Locally Deployed on Consumer-grade Hardware

Rui Xiao and Yili Xu

Whole genome sequencing (WGS) is essential for precision oncology, yet its clinical adoption remains limited by prohibitive computational costs and multi-day turnaround times. This work presents a fully localized low-resource framework enabling stable deployment of a trillion-parameter biomedical LLM on a single consumer-grade RTX 4060 laptop with 32GB system memory and 8GB VRAM, as well as on routine clinical workstations in general hospitals, completing the entire tumor-paired WGS workflow from raw FASTQ input to clinical-grade full-variation-spectrum report output. Under standard 30X depth configurations, our implementation finishes a single tumor-paired WGS analysis within 18 hours, achieving 99.62% F1 score for somatic variant detection with over 99.9% concordance to the industrial-standard A100 cluster pipeline, fully meeting clinical oncology accuracy requirements. Quantitative profiling shows adaptive heterogeneous memory scheduling accounts for 71% of total execution time, while model optimization introduces less than 9% of total detection error. This work is the first engineering implementation of trillion-parameter biomedical LLM-driven clinical-grade genomic analysis on consumer-grade hardware, breaking the industry paradigm that trillion-scale genomic LLMs require hundred-thousand-dollar GPU clusters and multi-day turnaround, establishing a low-resource pathway for global primary medical institutions to adopt whole-genome precision oncology at zero additional cost.
Read the full abstract

Genomics · Cross-listed from cs.AI

Towards a knowledge-enhanced single-cell foundation model

Hanqing Zhang, Jie Bao, Mei Ma et al. (+7 more)

Single-cell foundation models (scFMs) increasingly rely on large-scale transcriptomic pretraining, yet expanding pretraining data can yield diminishing gains while substantially increasing computational cost. Our data scaling analyses showed that incorporating biological knowledge, including cell-level text annotation and gene-level regulatory information, provided additional scaling dimension than simply increasing data size. Motivated by this observation, we present scKITE, a simple yet effective scFM that integrates cell-annotation and gene-regulatory supervision into a shared transcriptomic Transformer encoder through lightweight auxiliary decoders. These decoders are used only during pretraining and subsequently discarded, yielding a general-purpose encoder enriched with biological knowledge for downstream applications. With only 179,067 pretraining samples, i.e., less than 0.5% of those used by previous strong scFMs, scKITE outperformed these models across diverse downstream tasks, highlighting knowledge-enhanced pretraining as a promising paradigm for biologically grounded scFMs.
Read the full abstract

All authors Hanqing Zhang, Jie Bao, Mei Ma, Shuai Liu, Jiaying Ma, Jiaguan Liu, Jiaxiao Li, Zhenbo Li, Wenwen Gong, Zhijun Ca

Sunday 13 September

Genomics

Structure is not mechanism: high-gain gated-FFN rows across text and genomic foundation models

Alexandros Tzanakakis, Aris Karatzikos and Ilias Georgakopoulos-Soares

A small number of unusually high-gain parameters can exert disproportionate effects in transformer language models, but whether analogous structures recur in genomic foundation models and whether structural geometry determines functional importance remains unknown. We analyzed high-gain rows in gated feed-forward networks across text and genomic foundation models, including a frozen 22-model causal census. Computing an associated bilinear weight operator exactly, without a diagonal approximation, we tested whether structural extremeness is a transferable mechanism. Activation-derived candidates were functionally enriched relative to random and top-norm same-layer controls, yet neither spectral concentration nor operator magnitude predicted causal effect size, and these associations vanished within the endpoint-homogeneous text-decoder subset. A within-layer sweep of 36 rows in one genomic and one text decoder resolved this into two regimes: below the detector's acceptance threshold the ratio carried no positive information about causal damage, whereas above it the ratio ordered rows strongly but did not grade severity as a dose-response. The same sweep revealed a second individually catastrophic row invisible to a one-candidate-per-model census, and non-additive damage among co-located critical rows. Case studies showed divergent causal organizations: a robust super-additive pair interaction in DNABERT-2, and in GENERator a sharply position-localized dependence in which preserving or restoring the row's beginning-of-sequence contribution rescued essentially all native-loss damage. High-gain gated-FFN rows are therefore a recurrent architectural phenotype whose structural prominence acts as an enrichment signal, not a calibrated measure of functional criticality or a specification of causal organization. Enrichment is general, but the mechanism is model-specific.
Read the full abstract

Saturday 12 September

Genomics

RAGCell: Retrieval-Augmented Generation as Supervision for Versatile Single-cell Analysis

Tianyu Liu, Fan Zhang, Jiayuan Chen et al. (+10 more)

Single-cell foundation models (scFMs) are transforming computational biology by enabling generalizable, task-agnostic representations for versatile single-cell analysis. Despite their progress in facilitating rapid deployment for downstream tasks, off-the-shelf scFMs still have some overlooked concerns: (I) (Pretraining Cost.) Pretrain-based scFMs necessitate pretraining on a vast volume of cells, rendering it draining resources in applications. (II) (Heterogeneous Gap.) Large Language Models (LLM)-based scFMs ignore the tremendous heterogeneous gap between LLM textual and raw cellular spaces, leading to insufficient capability when facing downstream tasks. To this end, we introduce RAGCell, a versatile single-cell analysis framework that achieves a double-win in both cost-effectiveness and high performance. The success of RAGCell lies in two key aspects: Leveraging LLMs to construct cell-level and feature-level knowledge databases, which serve as supervision signals for training the cell model and significantly reduce the training cost ($>$pretrain-based scFMs). Aligning cell representations with text embeddings from the bi-level knowledge databases, enabling knowledge transfer from textual spaces to cellular spaces and effectively mitigating the heterogeneous gap ($>$LLM-based scFMs). Through extensive experiments on six downstream single-cell analysis tasks, we demonstrate that RAGCell achieves outstanding performance compared to state-of-the-art scFMs while operating at less than $\sim$1/10 the cost of pretrain-based scFMs.
Read the full abstract

All authors Tianyu Liu, Fan Zhang, Jiayuan Chen, Kun Wang, Haoxuan Li, Shengju Qian, Zhihong Zhu, Donghao Zhou, Hao Wu, Ziheng Zhang, Zhenxi Lin, Xian Wu, Yefeng Zheng

Thursday 10 September

Genomics · Cross-listed from q-bio.QM

Biology-in-the-loop: Amortized Adaptive Hit Discovery in CRISPR Screens

Carl Edwards, Edward De Brouwer, Xiner Li et al. (+5 more)

Many biological discovery problems require experiments to be selected sequentially under constrained budgets. CRISPR screening is a prominent example, as exhaustive perturbation testing is often infeasible and candidate perturbations must instead be prioritized over multiple experimental rounds. Despite the importance of this problem, existing benchmarks for adaptive hit discovery remain limited in scale and diversity. Here, we introduce AssayBench-Loop, a large-scale benchmark for adaptive hit discovery comprising 1,389 CRISPR screens across five phenotype categories. Beyond enabling systematic evaluation, its scale makes it possible to learn acquisition strategies across historical experiments. Building on this resource, we introduce AssayLoop, a sequential experimental design framework combining AssayFormer, a transformer-based amortized acquisition policy trained across historical screens to adapt from experimental feedback, with LLM-derived biological priors through an adaptive handoff. In this view, completed experiments become training data for learning how accumulated evidence should guide what to test next, while LLMs provide prior biological knowledge to seed the search. We further introduce AssayLLM, showing that the same principle can be extended directly to an LLM through task-specific post-training. On temporally held-out screens, AssayLoop achieves a 5.67-fold enrichment over random selection and recovers 27.7% of hits after assaying approximately 5% of the candidate library, outperforming existing adaptive-design methods and standalone LLMs, and AssayFormer alone. Performance improves with increasing historical training data and transfers to phenotype categories excluded from training. These results demonstrate the value of learning acquisition policies across historical experiments and combining them with broad biological priors for efficient adaptive hit discovery.
Read the full abstract

All authors Carl Edwards, Edward De Brouwer, Xiner Li, Namkyeong Lee, Ehsan Hajiramezanali, Anne Biton, Sara Mostafavi, Gabriele Scalia

Tuesday 8 September

Genomics

A Transformer-Based Delta Expression Encoder for Psilocybin Transcriptional Response: Architecture, Representations, and Biological Validation

Sai Jayakumar

Understanding why individuals respond differently to psilocybin requires modeling the drug's transcriptional perturbation signature at the cell-type level. I present a Transformer-based delta expression encoder that learns to classify differential gene expression status - upregulated, downregulated, or neutral - from single-nucleus RNA-sequencing data, without supervision from pathway annotations or prior biological knowledge. The model is trained on pseudobulk profiles from 623 examples spanning 18 cell types, 2 drug conditions, and 6 timepoints derived from the Liao et al. 2025 dataset, and achieves 69.4% weighted classification accuracy. Three principal findings are reported, alongside one direct test of a published hypothesis that returned a result inconsistent with that hypothesis. First, per-cell-type classification accuracy ranges from 28.3% (L2/3 IT, a primary HTR2A-expressing psilocybin target) to 99.6% (endothelial cells), consistent with known psilocybin response biology. Second, psilocybin-induced transcriptional downregulation is significantly more stereotyped across individuals than upregulation (Mann-Whitney U=18615.0, p<0.0001), a novel finding with a cortical depth gradient across excitatory subtypes. Third, attention-guided gene co-regulation analysis recovers drug-specific modules without pathway supervision. Separately, a direct test of whether baseline HTR2A expression predicts drug-response separability across cell types found a significant negative correlation (Spearman r = -0.7088, p = 0.0021), the opposite of what a simple HTR2A-gating account would predict.
Read the full abstract

Monday 7 September

Genomics

Human mutation field reveals an equilibrium-like structure with irreversible circulation

Isabella Caranzano, Daniel Maria Busiello, Stefano Priorelli et al. (+2 more)

The evolution of DNA sequences can be viewed as stochastic dynamics on a high-dimensional discrete space, but it is unclear when empirical transition biases reduce to an effective energy landscape versus retain irreducible non-equilibrium circulation. Human context-dependent mutation probabilities offer a direct test: every single-nucleotide substitution in a local context has a reverse substitution, so the logarithm of the forward-to-reverse probability ratio defines an antisymmetric field-the human mutation field. We show this field has a dominant gradient component and a smaller but reproducible curl component. Using seven-base human germline substitution probabilities, we infer an effective mutational landscape with a Siamese neural network constrained to predict only energy differences. This model predicts forward-to-reverse log-ratios for held-out mutations with a correlation of about 0.93, close to both an unconstrained predictive reference (0.948) and the empirical reversible ceiling from Hodge projection (about 0.96). Although trained only on mutation probabilities, the inferred landscape largely recovers short-word genomic composition and Chargaff reverse-complement symmetry for sequences up to length four. Deviations from equilibrium structure reveal a small but detectable nonequilibrium component: a residual irreversible circulation violating the Kolmogorov cycle condition for detailed balance, reproducible across African, Asian, and European populations, and strongest in CpG-linked cycles and CpG-transition edges, consistent with methylcytosine deamination. These results give a thermodynamic decomposition of the human mutation field: most mutation bias is organized by a local equilibrium-like energy landscape aligned with genome composition, while the residual circulation points to specific directional mutational mechanisms.
Read the full abstract

All authors Isabella Caranzano, Daniel Maria Busiello, Stefano Priorelli, Amos Maritan, Piero Fariselli

Saturday 5 September

Genomics · Cross-listed from stat.AP

A Network-Structured Bayesian Hierarchical Model for Sparse Mutation-Drug Response Associations: Application to Cancer Pharmacogenomics

Hammed A. Olayinka and Saheed O. Olayemi

We develop a network-structured Bayesian hierarchical model for sparse association mapping between genomic alterations and quantitative treatment-response phenotypes. The framework combines a Gaussian Markov random field prior that borrows strength across pathway-connected genes, a global-local horseshoe prior inducing sparsity, and a conjugate Gibbs sampler requiring no Metropolis-Hastings steps. Though broadly applicable to high-dimensional settings with known predictor networks, we validate it using cancer cell-line drug-sensitivity data. Applied to GDSC2 ($N=951$ cell lines, $G=219$ driver genes, $D=295$ drugs), the model identifies 126 gene-drug associations (0.195% of 64{,}605 pairs), concentrated in EZH2 (45 drugs, all sensitivity-direction, mean effect $-0.911$ $\ln$IC50) and KMT2D (36 drugs, all sensitivity-direction, mean effect $-0.496$ $\ln$IC50). These markers show external support in an independent PRISM screen (1{,}518 compounds), with KMT2D achieving complete directional replication (36/36) and EZH2 partial replication (8/12). Five-fold cross-validated predictive log-likelihood confirms each prior layer's value: the full model outperforms the no-network ablation by $+3{,}109$ log-units per fold and the no-horseshoe ablation by $+14{,}039$ log-units, consistently across folds. Simulations under three scenarios show the full model achieves the highest precision and lowest false-discovery rate throughout, while the network prior improves sensitivity recovery under network-structured signal. A tissue-stratified extension identifies coherent subgroup refinements, including lung-specific EGFR-inhibitor sensitivity and skin-specific BRAF-Dabrafenib sensitivity. These results show the framework identifies sparse, interpretable, externally supported drug-sensitivity markers while enabling principled investigation of tissue-specific departures from shared effects.
Read the full abstract

Friday 4 September

Genomics · Cross-listed from cs.IR

VizIt: A multi-view framework for exploring single-cell, spatial, and genetic data online

Chenhang Christopher Zhang, Yanqing Lou, Jie Yuan et al. (+8 more)

Multi-omic studies increasingly require data to be examined from complementary biological perspectives, yet interactive exploration remains fragmented across modalities and tools. We present VizIt, an open-source framework for multi-view exploration of single-cell and spatial transcriptomic, epigenomic and genetic data. VizIt connects gene-, cell type-, condition-, spatial-, genomic region- and variant-centered views, enabling seamless navigation across biological perspectives. We demonstrate VizIt through the Parkinson's Cell Atlas, a customizable interactive multi-omic resource.
Read the full abstract

All authors Chenhang Christopher Zhang, Yanqing Lou, Jie Yuan, Mingming Lu, Jacob Parker, Himanshu Chintalapudi, Zechuan Lin, Clemens R. Scherzer, Yuxuan Hu, Ruifeng Hu, Xianjun Dong

Thursday 3 September

Genomics · Cross-listed from cs.DB

A Semantic Model of Genetic Evidence: A Step Toward Bridging the Basic-Science-Clinic Gap

Michael Bouzinier and Dmitry Etin

Scientific and clinical decision-making depends on evidence from the primary literature, but existing standards for representing that evidence (FHIR Evidence, ECO, SEPIO, and the GA4GH Genomic Knowledge Standards) are oriented toward clinical-trial workflows, evidence codes, or single-variant assertions, and do not capture the fine-grained, domain-specific structure of claims in basic and pre-clinical research. We introduce a semantic model for scientific evidence with three core classes, specialize it for genetics, align it structurally to FHIR Evidence with a SEPIO-anchored credibility decomposition, and attach a compact dimensional vocabulary whose conditional-activation rules are validated by a SHACL schema for the implemented constraints. Using clinical variant interpretation as the driving use case, we evaluate the model through a human-AI annotation pilot over six genetics papers, yielding 28 evidence items and 95 source-anchored assertions, with a workflow that keeps curator-authored reference annotations distinct from AI-drafted annotations. Treating the pilot as a feasibility study rather than a benchmark, we argue that the model is a useful increment toward trustworthy, AI-ready infrastructure for variant interpretation: a reference data model and validation schema for representing genetic evidence.
Read the full abstract

Wednesday 2 September

Genomics

Enhancer-promoter proximity predicts transcriptional competence but not transcriptional output in the Drosophila brain

Olivier Messina, Loucif Remini, Christopher H. Bohrer et al. (+4 more)

How 3D genome architecture contributes to transcriptional specificity across neuronal cell types remains unclear. Here, we used multiplexed chromatin tracing to map chromatin architecture and cell identity at single-cell resolution in the adult Drosophila brain. We found that enhancer-promoter (E-P) proximity was increased in transcriptionally active compared with inactive neurons. Analysis of single traces revealed the existence of distinct proximal and distal E-P states, with active neurons enriched in the proximal state. However, this relationship broke down across active neuronal subtypes, where neither E-P proximity nor chromatin accessibility predicted transcriptional output. Thus, 3D genome organization distinguishes transcriptionally competent from inactive neuronal states without quantitatively specifying transcriptional output. Our findings support a model in which E-P proximity establishes a permissive structural state, while additional cell-type-specific regulatory mechanisms tune transcriptional output.
Read the full abstract

All authors Olivier Messina, Loucif Remini, Christopher H. Bohrer, Jean-Bernard Fiche, Jean-Charles Walter, Andrea Parmeggiani, Marcelo Nollmann

Genomics

Subcellularly Resolved Single-Cell Embedding Learning with Transcriptomic data, Protein Structure and Localization Information

Zhen Zhou, Jiachen Li, Yuan Liu et al. (+2 more)

Existing cell embedding methods predominantly rely on transcriptomic or proteomic measurements and represent each cell as a holistic entity, thereby overlooking the subcellular localization of individual molecules. Moreover, they rarely incorporate protein structural information, despite its fundamental role in determining molecular interactions and functions. In this work, we propose a multimodal framework for learning subcellularly resolved cell embeddings by jointly leveraging RNA expression profiles, protein sequence representations, and protein structural information. Specifically, we employ a cross-attention architecture to integrate transcriptomic, sequence, and structural modalities and model their interactions within distinct subcellular compartments. The resulting embeddings represent each cell through its fine-grained subcellular organization, capturing both molecular expression patterns and the functional properties of the associated proteins. By learning cell representations at subcellular resolution, our framework preserves spatially organized biological information while integrating complementary signals across multiple molecular levels. To the best of our knowledge, this is the first framework that produces subcellularly resolved cell embeddings by jointly incorporating transcriptomic information, protein sequence representations, and protein structural knowledge within a unified cross-modal learning paradigm.
Read the full abstract

All authors Zhen Zhou, Jiachen Li, Yuan Liu, Xiaoyong Pan, Hong-Bin Shen

Tuesday 1 September

Genomics

PopPert: Population-level Joint-Distribution Modeling for Single-Cell Perturbation Prediction

Handong Wang, Jiaxin Qi, Haochen Feng et al. (+1 more)

Predicting transcriptional responses to specific perturbations is critical for understanding cellular regulatory mechanisms and accelerating drug discovery. Single-cell RNA sequencing destroys each measured cell, yielding only unpaired populations of control and perturbed cells. However, existing methods typically model perturbation prediction at the single-cell level and assume cell-to-cell correspondence, which conflicts with the unpaired nature of the observed data. To address this challenge, we propose PopPert, a framework that explicitly parameterizes population-level joint gene expression distributions for collective transcriptional state modeling. Given a control population distribution and a perturbation condition, PopPert predicts perturbation-induced changes in distribution parameters, eliminating the need for cell-level correspondence and reducing sensitivity to single-cell noise. To effectively capture gene co-expression patterns, PopPert leverages a low-rank Gaussian Copula to model cross-gene statistical dependencies and construct the joint gene expression distribution, additionally allowing sampling of synthetic perturbed single-cell profiles. Across multiple single-cell benchmarks spanning both genetic and chemical perturbations, PopPert achieves superior overall performance in differential expression recovery, perturbation effect estimation, and population-level distribution matching. These results establish population-level joint distribution learning as an effective paradigm for predicting transcriptional responses from unpaired single-cell populations. Code for PopPert is publicly available at https://github.com/whd1125/PopPert.
Read the full abstract

All authors Handong Wang, Jiaxin Qi, Haochen Feng, Baisheng Lai