SEARCH · Search PubMed

Results for “Protein Language Model”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMed2026

PepGen: conditional generation of peptides for MHC binding.

MOTIVATION: Peptide-MHC II binding drives adaptive immunity, yet discovery of novel binder peptides remains challenging due to open binding grooves of MHC-II that accommodate variable-length peptides. While discriminative models perform well, they are unfeasible for generation via enumeration due to vast peptide space (2013≈8×1016 for peptides of length 13 amino acids). Generative AI approaches could accelerate binder design to enable vaccines targeted to particular MHC-II alleles or optimize other peptide chemical properties. RESULTS: We introduce PepGen, the first protein language model for MHC II peptide generation building on Generalized Language Modeling. PepGen conditions on alleles, arbitrary partial peptides including putative TCR-interacting motifs, and continuous binding affinity. Across multiple benchmarks including infilling and de novo generation, PepGen outperformed frequency sampling, Gibbs clustering, and autoregressive baselines. Adjusted log-probabilities enable good classification performance. Experimental validation confirmed that the SARS-CoV-2 peptide TEGALNTPKDHIGTR binding the HLA-DQA101:03-DQB106:03 allele can be redesigned to bind the HLA-DQA101:02-DQB105:02 allele. PepGen generated three putative TCR-motif-preserving binders gaining up to 70% of original MFI. Overall, PepGen provides scalable, motif-constrained MHC II peptide redesign and de novo generation, validated through thorough benchmarks and functional assays. AVAILABILITY AND IMPLEMENTATION: Code and Data are available at https://github.com/DaniTheOrange/PepGen.

Peptides↗

PubMed2026

FANTASIA suite: a reproducible and configurable framework for embedding-based functional annotation of proteins.

Embedding-based annotation transfer is increasingly used for protein function inference due to protein language models capture sequence, structural, and functional signals that may extend beyond conventional pairwise similarity. However, systematic application of these approaches requires control over model choice, reference composition, lookup parameters, evidence traceability, and output formats. We developed the FANTASIA suite, a configurable framework for embedding-based functional annotation of proteins. The suite combines a database-backed implementation for reproducible and extensible analyses with a portable flat-file implementation for rapid local annotation and pipeline integration. Using non-model and model-organism proteomes, we show that larger neighbourhood sizes remain practical for proteome-scale analyses and that taxonomy and sequence-identity filtering support leakage-aware benchmarking. We also compare the supported models with baseline methods through external CAFA5 evaluation and provide practical guidance based on empirical evidence variables. FANTASIA provides a controlled, scalable, and reproducible framework for extending functional annotation across the rapidly expanding diversity of sequenced organisms.

Software↗

PubMed2026

EscaPRRS-ORF5: a structure-aware evolutionary framework for prioritizing immune escape-prone variants in porcine reproductive and respiratory syndrome virus.

MOTIVATION: Porcine Reproductive and Respiratory Syndrome Virus (PRRSV) is a rapidly evolving RNA virus causing significant economic losses, posing a formidable challenge to vaccine efficacy due to its high mutational variability and immune escape. As the viral mutants evolve, their ability to sustain in population is driven by a range of host biology factors such as receptor binding, fusion, and uncoating. Existing tools that predict viral fitness and escape propensities rely heavily on extensive, up-to-date sequence data and lack integration of biochemical host interactions, limiting mechanistic understanding of the mutational landscape. We introduce Esca, a sequence-only toolchain framework that identifies immune escape-prone residues by exhaustively scanning each residue position for all amino acid substitutions using a Bayesian Variational Autoencoder (VAE) trained on protein language model embeddings. We demonstrate Esca on the GP5(ORF5) glycoprotein of PRRSV (EscaPRRS-ORF5) by training on ESM-2 embeddings of 32 146 GP5 sequences (2015-2022) spanning 140 sub-lineages. RESULTS: Despite being trained only on GP5 sequence data, EscaPRRS-ORF5 recovered 85.7% of the surface-exposed receptor binding interfaces as escape-prone regions. We use a mutation-sensitive fitness scoring scheme that goes beyond Hamming distances, to predict antibody escape tendencies, supporting surveillance of (re) emerging PRRSV variants. We do not claim that ORF5 alone captures PRRSV evolution or serves as a surveillance endpoint; rather, Esca offers a scalable path toward whole-genome, structure-aware surveillance. AVAILABILITY AND IMPLEMENTATION: EscaPRRS-ORF5 is freely available at https://doi.org/10.6084/m9.figshare.32661033 with an interactive Colab notebook at https://colab.research.google.com/drive/1TEgzAhPwvNAZ01VXeJbIFibfri2jnDA5? usp=sharing.

Porcine respiratory and reproductive syndrome viru↗

PubMed2026

Fine-grained structural classification of biosynthetic gene cluster-encoded products.

MOTIVATION: Biosynthetic gene clusters (BGCs) are responsible the biosynthesis of many natural products, including a multitude of effective therapeutics and their precursors. Advances in genomic data collection as well as computational techniques have made it possible to identify BGCs at scale. However, accurately determining the types of BGC-encoded products from genomic content remains elusive. RESULTS: Here, we introduce BGC annotation tool (BGCat), a machine learning method for fine-grained structural classification of BGC-encoded products, leveraging the NPClassifier natural product nomenclature. Our method leverages a pre-trained protein language model for creating meaningful gene representations and a deep neural network for class label prediction. We show the method outperforms state-of-the-art approaches in coarse-grained product classification and is effective for detailed classification. We implement a clustering-based augmentation strategy for BGC-product relationships, addressing a crucial gap in the available datasets. We then introduce the concept of product class profiles of gene cluster families (GCFs), associating each GCF with a probabilistic distribution of product types and offering a new perspective on GCF functions. Lastly, we use BGCat to provide new product class labels for over 100k BGCs in antiSMASH DB that presently have minimal information about their products. AVAILABILITY AND IMPLEMENTATION: The source code and trained model weights are freely available at https://github.com/HassounLab/BGCat.

Multigene Family↗

PubMed2025

mamp-ml: A deep learning approach to epitope immunogenicity in plants.

Eukaryotes detect biomolecules through surface-localized receptors, key signaling components. A subset of receptors survey for pathogens, induce immunity, and restrict pathogen growth. Comparative genomics of both hosts and pathogens has unveiled vast sequence variation in receptors and potential ligands, creating an experimental bottleneck. We have developed mamp-ml, a machine learning framework for predicting plant receptor-ligand interactions. We leveraged existing functional data from over two decades of foundational research, together with the large protein language model ESM-2, to build a pipeline and model that predicts immunogenic outcomes using a combination of receptor-ligand features. Our model achieves 73% prediction accuracy on a held-out test set, even when an experimental structure is lacking. Our approach enables high-throughput screening of LRR receptor-ligand combinations and provides a computational framework for engineering plant immune systems.

Journal Article↗

PubMed2026

Rational and computation-assisted engineering of a compact and efficient CRISPR-Cas12f genome editor.

The CRISPR-Cas12f system is an ultracompact genome-editing platform, yet only a few orthologs exhibit robust activity in mammalian cells. Here, we systematically screened 23 Cas12f orthologs and identified two active nucleases, PspCas12f1 and TcCas12f1, capable of genome editing in human cells. Single guide RNA (sgRNA) scaffold optimization enhanced the basal activity of PspCas12f1. To further improve its performance, we combined structure-guided rational design with protein language model-assisted filtering. Candidate mutations predicted by SaProt were further screened based on structural proximity to the DNA-binding interface and electrostatic compatibility. This integrative strategy identified Q100R and E293R, whose combination yielded the optimized variant enPspCas12f1. enPspCas12f1 achieved genome-editing efficiencies comparable to SpCas9 across multiple endogenous loci while maintaining high specificity. Collectively, our results demonstrate that integrating protein language model-assisted filtering with structure-guided rational design provides an effective strategy for engineering PspCas12f1 and may facilitate the optimization of additional compact CRISPR nucleases.

CRISPR-Cas12f↗

PubMed2025

CaLMPhosKAN: prediction of general phosphorylation sites in proteins via fusion of codon aware embeddings with amino acid aware embeddings and wavelet-based Kolmogorov-Arnold network.

MOTIVATION: The mapping from codon to amino acid is surjective due to codon degeneracy, suggesting that codon space might harbor higher information content. Embeddings from the codon language model have recently demonstrated success in various protein downstream tasks. However, predictive models for residue-level tasks such as phosphorylation sites, arguably the most studied Post-Translational Modification (PTM), and PTM sites prediction in general, have predominantly relied on representations in amino acid space. RESULTS: We introduce a novel approach for predicting phosphorylation sites by utilizing codon-level information through embeddings from the codon adaptation language model (CaLM), trained on protein-coding DNA sequences. Protein sequences are first reverse-translated into reliable coding sequences by mapping UniProt sequences to their corresponding NCBI reference sequences and extracting the exact coding sequences from their GenBank format using a dynamic programming-based global pairwise alignment. The resulting coding sequences are encoded using the CaLM encoder to generate codon-aware embeddings, which are subsequently integrated with amino acid-aware embeddings obtained from a protein language model, through an early fusion strategy. Next, a window-level representation of the site of interest, retaining the full sequence context, is constructed from the fused embeddings. A ConvBiGRU network extracts feature maps that capture spatiotemporal correlations between proximal residues within the window. This is followed by a prediction head based on a Kolmogorov-Arnold network (KAN) using the derivative of gaussian wavelet transform to generate the inference for the site. The overall model, dubbed CaLMPhosKAN, performs better than the existing approaches across multiple datasets. AVAILABILITY AND IMPLEMENTATION: CaLMPhosKAN is publicly available at https://github.com/KCLabMTU/CaLMPhosKAN.

Codon↗

PubMed2025

Redefining ALS: Large-scale proteomic profiling reveals a prolonged pre-diagnostic phase with immune, muscular, metabolic, and brain involvement.

BACKGROUND: Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disorder with a largely unknown duration and pathophysiology of the pre-diagnostic phase, especially for the common non-monogenic form. METHODS: We leveraged the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort with up to 30 years of follow-up to identify incident ALS cases across five European countries. Pre-diagnostic plasma samples from initially healthy participants underwent high-throughput proteomic profiling (7,285 protein markers, SomaScan). Cox proportional hazards models based on 4,567 participants (including 172 incident ALS cases) were used to identify protein biomarkers associated with future ALS diagnosis. Top results were indirectly validated in two independent case-control studies of prevalent ALS (n=417 ALS, 852 controls). Functional annotation included cross-disease comparisons, gene set and tissue enrichment testing, organ-specific proteomic clocks, and the application of large-language models (LLM). FINDINGS: Five proteins (SECTM1, CA3, THAP4, KLHL41, SLC26A7) were identified as significant pre-diagnostic ALS biomarkers (FDR=0.05), detectable approximately two decades before diagnosis. Of these, all except SECTM1 were indirectly validated in independent cohorts of prevalent ALS cases, supporting their clinical significance. Additionally, 22 nominally significant (p<0.05) pre-diagnostic biomarkers were FDR-significant in prevalent ALS with consistent effect directions. Cross-disease comparisons with pre-diagnostic Parkinson's and Alzheimer's disease suggested a largely specific pre-diagnostic ALS biomarker signature. Gene ontology and tissue enrichment highlighted early involvement of immune, muscle, metabolic, and digestive processes. Furthermore, analyses of proteomic clocks revealed accelerated aging in brain-cognition, immune, and muscle tissues before clinical diagnosis. Druggability and LLM analyses revealed possible therapeutic targets and novel strategies, emphasizing translational relevance. INTERPRETATION: Our study provides first evidence of ultra-early molecular changes in common ALS up to two decades prior to clinical onset, mainly affecting immune, muscle, metabolic, digestive, and cognitive systems. Our study nominates several compelling candidates for risk stratification studies and novel therapeutic targets for early intervention. FUNDING: Clinical Research in ALS and Related Disorders for Therapeutic Development (CreATe) Consortium, Cure Alzheimer's Fund, Michael J Fox Foundation, Interdisciplinary Centre for Clinical Research, University Münster.

Journal Article↗

PubMed2025

Redefining ALS: Large-scale proteomic profiling reveals a prolonged pre-diagnostic phase with immune, muscular, metabolic, and brain involvement.

Journal Article↗

PubMed2026

Large language models in bioinformatics: a comprehensive survey.

The emergence of foundation models with trillion-level parameters has redefined the landscape of artificial intelligence. Various fields are developing their own large-scale models, which can solve many problems within the field and improve work efficiency. Biological large-scale models are a cross-disciplinary research field that combines mathematics, computer science, and biology, aiming to simulate and understand the structure, function, and dynamic changes of biological systems through the establishment of complex computational models. This field covers multiple levels such as biological pathways, population dynamics, protein folding, etc., providing us with tools for deep exploration of the mysteries of life and applications in medicine, ecology, and other fields. This article reviews the background and research status of biological large-scale models, and discusses future directions. Large language models (LLMs) and other large-scale foundation models have rapidly advanced in recent years, enabling powerful representation learning and generation across text, sequences, and multimodal data. In bioinformatics and biomedicine, these models are increasingly used to analyze genomic sequences, infer protein properties and structures, support drug discovery, and integrate heterogeneous biomedical evidence. This survey reviews the basic principles of LLMs and summarizes representative applications in (i) gene and genome sequence analysis, (ii) protein structure and function prediction, and (iii) drug design, including virtual screening and personalized medicine. We also discuss emerging multi-model modeling approaches, as well as key challenges such as data quality and privacy, interpretability, generalization to new organisms and tasks, and responsible deployment in health-related settings. Finally, we outline future directions for developing reliable, scalable, and explainable bioinformatics foundation models.

bioinformatics↗

PubMed2026

Accelerating inference in genomic and proteomic foundation models via speculative decoding.

MOTIVATION: Genomic and protein foundation models (GFMs and PFMs) have demonstrated strong performance in learning the language of DNA and proteins, but their use in large-scale sequence generation is limited by the latency of autoregressive decoding. Because every token triggers a forward pass of a large Transformer, whose inference is relatively slow, long-sequence generation quickly becomes costly. RESULTS: In this work we adapt speculative decoding to a representative GFM: the DNA model DNAGPT and two representative PFMs: ProGen2 and ProtGPT2. We implement a probabilistic variant of speculative decoding, in which a lightweight draft model proposes short token spans and a larger target model verifies or corrects them in parallel, while preserving the target model's sampling distribution. Across all three models we systematically study the effect of speculation window length, temperature, draft architecture and prompt length, and we benchmark tokens per second over multiple runs per configuration. Speculative decoding yields consistent speedups over standard key-value cached decoding, with maximum observed speedup reaching 100% increase, while average gains across models ranging between 20% and 40% (e.g. 1.2×-1.4×), without changing the underlying target model predictions. Our results show that speculative decoding is a practical and model-agnostic strategy for accelerating genomic and proteomic sequence generation without sacrificing prediction quality. AVAILABILITY AND IMPLEMENTATION: All code and results are freely available at https://github.com/Georgakopoulos-Soares-lab/BioSpecDec.

Genomics↗

PubMed2026

A trainable language model with potential to modulate translation rates in non-model organisms by generating upstream untranslated region sequence libraries.

Tuning protein expression in non-model organisms is often constrained by the lack of validated genetic parts and predictive design tools. Translational tuning through the modulation of upstream untranslated regions (5'-UTRs) offers a potentially organism-agnostic route, but existing methods typically rely on mechanistic assumptions, prior knowledge that may not be available in non-model contexts, or the screening of sequence libraries. Here, we present a simple generative approach for creating synthetic 5'-UTR libraries based solely on the genomic sequence statistics of any desired organism. The method uses a sliding-window n-gram language model applied to native 5'-UTR sequences to produce novel sequences that preserve organism-specific base distributions and motifs without hard-coding specific motifs or mechanistic rules into inflexible statistical templates. We have applied this approach to the model bacterium Escherichia coli and the non-model probiotic Limosilactobacillus reuteri. Libraries of approximately 1,000 sequences were generated for each organism, from which about 100 unique sequences were experimentally tested for translation of a fluorescent reporter protein. In both organisms, the synthetic libraries yielded a broad range of translation levels from this relatively small number of tested variants. Sequences derived from an organism's own genomic statistics provided a more uniformly distributed range of translation rates in that organism than sequences derived from the other species. Correlations of individual sequence performance across the two species were weak, and thermodynamic predictions of ribosome binding strength showed very little predictive power, especially in the non-model L. reuteri. The results demonstrate that simple statistical language model approaches applied to genomic data can generate functional translational regulatory sequence libraries without detailed mechanistic knowledge or explicit reference to consensus motifs. The approach requires minimal computational resources, avoids reproducing native sequences, and can be readily applied to any organism with a sequenced genome. This strategy may lower technical barriers to expression tuning in non-model organisms.

5' Untranslated Regions↗

PubMed2026

Artificial Intelligence for Natural Products Discovery and Development.

Natural products (NPs) remain a cornerstone of modern drug discovery, offering stereochemical complexity and diverse bioactivities that precisely modulate therapeutic targets, refined through billions of years of evolution. However, their research has long been hindered by inefficient, empirical workflows, high resource consumption, structural complexity, and the "multicomponent, multi-target" nature of their mechanisms. The exponential growth of genomic, metabolomic, and spectral data has overwhelmed conventional analytical methods, exposing critical bottlenecks in handling high-dimensional, heterogeneous datasets that exceed human interpretive capacity. Artificial intelligence (AI) is emerging as a transformative paradigm to address these challenges, integrating multi-omics and chemical data to shift NP research from fragmented empiricism toward mechanism-driven, precision-oriented development. By leveraging deep learning architectures- including graph neural networks, Transformers, and diffusion-based generative models-AI enables systematic decoding of NP biosynthesis, automated structure elucidation, rational target identification, knowledge extraction from vast unstructured scientific literature, and de novo molecular design. This review comprehensively surveys recent advances in AI applications across the full NP discovery and development pipeline, encompassing genome mining, structure-based and ligand-based virtual screening, multimodal structural characterization, lead optimization, and biosynthetic pathway engineering. We further examine the emerging roles of protein-centric, molecule- centric, and multimodal foundation models, as well as large language models, in bridging genotype-to-chemotype gaps and unlocking unstructured scientific knowledge. Finally, we discuss critical challenges including data scarcity, representational limitations for complex stereochemistry, physical plausibility in generative models, and the urgent need for experimental validation, while outlining future directions toward autonomous experimentation, closed-loop optimization, and human-AI collaborative discovery.

Artificial intelligence↗

PubMed2026

Deep generative models in biological sequence and structure analysis and design.

Deep generative models have transformed biological sequence modeling from predictive analysis toward increasingly controllable design. Early biological applications of Variational Autoencoders (VAEs) and Generative Adversarial Networks (GANs) established latent representation learning and sequence synthesis, while recent advances in transformer-based language models, discrete diffusion, flow-matching, and multimodal generative frameworks have substantially expanded the scope of biological design. This review examines generative models for DNA, RNA, and protein sequence design, emphasizing how different model classes represent biological constraints, operate over discrete and continuous spaces, and integrate sequence, structure, and function. We compare VAEs, GANs, autoregressive and masked language models, diffusion models, and flow-based approaches across genomics, transcriptomics, and proteomics, with particular attention to controllability, long-range dependency modeling, structural grounding, generalization, and experimental utility. We further examine evaluation strategies, out-of-distribution generalization, and closed-loop design-build-test-learn workflows that connect in silico generation with empirical validation. We distinguish fundamental modality-dependent constraints including sequence discreteness, context length, structural coupling, and physical or thermodynamic requirements from architecture-dependent advantages that reflect the current state of the field. Current studies suggest that long-context models are particularly useful for genome-scale representation and sequence modeling, whereas structure-aware diffusion, flow-based, and inverse-folding approaches provide better frameworks for geometry-constrained RNA and protein design. This perspective provides a critical framework for understanding the present capabilities, limitations, and convergence of generative approaches toward reliable and experimentally grounded biological design.

Biological sequence analysis↗

PubMed2026

Plasma von Willebrand Factor and ADAMTS13 Interact With APOE-ε4 in Predicting Longitudinal Brain Atrophy and Cognitive Decline Over a 9-Year Follow-Up.

BACKGROUND: Von Willebrand factor (VWF) and ADAMTS13 (a disintegrin and metalloproteinase with thrombospondin type 1 motif, 13) are linked to dementia risk, and limited evidence suggests apolipoprotein E (APOE)-ε4 alters VWF release. This study assessed whether baseline VWF and ADAMTS13 levels predict neurodegeneration and cognitive decline and evaluated effect modification by APOE-ε4 carriership. METHODS: Vanderbilt Memory and Aging Project cohort participants (n=332, 73±7 years, 59% male) completed serial blood draw, neuropsychological assessment, and brain magnetic resonance imaging over 6.4 years (range 1.4-9.7 years). Baseline plasma VWF and ADAMTS13 levels were quantified using mass spectrometry and Olink. Fully adjusted linear mixed-effects models related protein×time and protein×APOE-ε4×time interaction terms to longitudinal brain magnetic resonance imaging and neuropsychological outcomes. RESULTS: Lower baseline ADAMTS13 predicted faster declines in language (β=0.11, P=0.01), information processing speed (β=0.27, P=0.001), executive function (β=0.01, P=0.03), episodic memory (β=0.01, P=0.03), and visuospatial ability (β=0.11, P=0.001) and faster increases in global (β=-0.29, P=0.01) and frontal (β=-0.17, P=0.01) white matter hyperintensity volumes. Associations between ADAMTS13 and faster rates of cognitive decline and white matter injury were driven by APOE-ε4 carriers. Models relating VWF to longitudinal outcomes were null. APOE-ε4 interacted with VWF on longitudinal gray matter volumetric outcomes, such that faster rates of global gray matter atrophy were observed with higher baseline VWF levels among APOE-ε4 noncarriers only (β=-1530.5, P<0.001). CONCLUSIONS: ADAMTS13 shows promise as a potential plasma biomarker for brain aging outcomes, but additional research is warranted to understand the performance of VWF in the presence versus absence of an APOE-ε4 allele.

Humans↗

PubMed2026

Post-translational modification of proteins in the human testis development pathway.

BACKGROUND: The foetal testes produce the androgens necessary to masculinise the developing embryo and support the maturation of germ cells, that will eventually develop into sperm, thus ensuring future reproductive capacity. The testes develop from the bi-potential gonads in a highly orchestrated process resulting in the differentiation of a complex tissue with multiple cellular lineages. While recent transcriptomic and chromatin-based analyses of human foetal testes have provided an unprecedented level of insight into signalling pathways activated during this process, proteomic studies of the human foetal gonads remain limited. Proteins are active molecules and post-translational modification (PTM) of proteins influences protein activity, stability and localisation. Studies have shown that PTMs regulate critical proteins in testis development, and their disruptions are implicated in congenital disorders including differences of sex development (DSD), in which sex development is atypical. Despite this, the role and regulation of protein PTM during human testis development remains poorly understood due to limited access to human foetal gonadal tissue, a paucity of large-scale proteomics studies, and a lack of robust of human gonad in vitro models. OBJECTIVE AND RATIONALE: This review aims to provide a comprehensive analysis of validated PTMs affecting proteins critical for testicular development. We discuss PTMs with evidence for a role in normal testis development, and highlight those disrupted in DSD. We review emerging techniques, including proteomic technologies and organ modelling systems that may advance our understanding of PTMs in foetal testis development. We discuss challenges that have restricted the application of these technologies and how overcoming these will significantly improve our understanding of testis development and disease, diagnostics and patient outcomes. SEARCH METHODS: We searched PubMed and the University of Melbourne library for peer-reviewed English-language studies using keywords such as phosphorylation, SUMOylation, acetylation, ubiquitination alongside each protein of interest. PTM sites in proteins involved in testis development were identified using the PhosphoSitePlus database focusing those confirmed in in vitro or animal model studies. ClinVar and the Human Gene Mutation Database were used to identify patient variants that may disrupt PTM sites. OUTCOMES: Our review finds that proteins required for human foetal testis development are subject to extensive PTM. Several PTM sites and PTM-mediated pathways [e.g. MAPK (mitogen-activated protein kinase) pathway] are disrupted in patients with DSD or related conditions. While recent advances in proteomics technologies hold considerable promise, their application to human foetal gonads has been constrained by technical, ethical, and logistical challenges. Encouragingly, emerging high-sensitivity and low-input technologies, alongside stem cell-based approaches, offer viable pathways to overcoming these barriers. WIDER IMPLICATIONS: The relationship between gene regulation, protein expression, and cellular outcome is inherently non-linear, shaped by additional regulatory layers-most notably PTMs. The contribution of PTMs to human testis development in both typical and atypical contexts is a major knowledge gap. Addressing this gap has broad clinical and biological relevance: it may help improve genetic diagnosis or shed light on how proteins or pathways critical for testis development respond to environmental signals-an increasingly pressing question as declining global fertility rates bring testicular function under greater scrutiny. REGISTRATION NUMBER: N/A.

Humans↗

PubMed2026

HyLnc: a hybrid deep learning and feature-based approach for long non-coding RNA prediction.

Long non-coding RNAs (lncRNAs) play important roles in gene regulation, development and disease, yet accurate identification of lncRNAs from transcriptomic data remains a major computational challenge. Existing methods often rely either on handcrafted sequence features or deep learning approaches, each with their inherent limitations in capturing the full complexity of RNA sequences. In this study, we proposed HyLnc, a computational framework that integrates transformer-based contextual embeddings with biologically meaningful sequence features for improved lncRNA prediction. A custom BERT-based model was first pre-trained on a large corpus of metazoan RNA sequences using a masked language modelling strategy to learn contextual nucleotide dependencies. The model was subsequently fine-tuned on curated datasets of lncRNAs and protein-coding transcripts and 256-dimensional deep sequence embeddings were extracted. Parallelly, 348 handcrafted features, including ORF characteristics, untranslated region (UTR) properties, nucleotide composition and Fickett scores, were computed. A multi-stage feature selection strategy was applied to identify the most informative features, resulting in optimized hybrid feature sets. Multiple machine learning classifiers were evaluated, with the RF model achieving the best performance. The proposed framework attained an accuracy of 91.30%, F1-score of 91.23% and MCC of 82.60 on an independent validation dataset, outperforming several existing lncRNA prediction tools. Thus, HyLnc demonstrates that integrating deep contextual representations with biologically interpretable features enhances lncRNA prediction. This approach provides a robust and scalable solution for large-scale transcriptome annotation and can be extended to other sequence-based prediction.

RNA, Long Noncoding↗

PubMed2025

Out-of-the-box bioinformatics capabilities of large language models (LLMs).

Large Language Models (LLMs), AI agents and co-scientists promise to accelerate scientific discovery across fields ranging from chemistry to biology. Bioinformatics- the analysis of DNA, RNA and protein sequences plays a crucial role in biological research and is especially amenable to AI-driven automation given its computational nature. Here, we assess the bioinformatics capabilities of three popular general-purpose LLMs on a set of tasks covering basic analytical questions that include code writing and multi-step reasoning in the domain. Utilizing questions from Rosalind, a bioinformatics educational platform, we compare the performance of the LLMs vs. humans on 104 questions undertaken by 110 to 68,760 individuals globally. GPT-3.5 provided correct answers for 59/104 (58%) questions, while Llama-3-70B and GPT-4o answered 49/104 (47%) correctly. GPT-3.5 was the best performing in most categories, followed by Llama-3-70B and then GPT-4o. 71% of the questions were correctly answered by at least one LLM. The best performing categories included DNA analysis, while the worst performing were sequence alignment/comparative genomics and genome assembly. Overall, LLMs performance mirrored that of humans with lower performance in tasks in which humans had low performance and vice versa. However, LLMs also failed in some instances where most humans were correct and, in a few cases, LLMs excelled where most humans failed. To the best of our knowledge, this presents the first assessment of general purpose LLMs on basic bioinformatics tasks in distinct areas relative to the performance of hundreds to thousands of humans. LLMs provide correct answers to several questions that require use of biological knowledge, reasoning, statistical analysis and computer code.

Journal Article↗