Biology subjects

Horlacher, M.

Publications and source records attributed to Horlacher, M..

8 recordsLinked to original sources

bioRxiv2026

PARNET: A CLIP-SEQ-BASED FOUNDATION MODEL FOR RNA SEQUENCE REPRESENTATION LEARNING

RNA-binding proteins (RBPs) orchestrate a complex combinatorial regulatory "code" that governs RNA splicing, stability, localization, and translation. Learning the relationship between RNA sequences and these processes is a central challenge in genomics. Foundation models, notably RNA language models, have emerged as the dominant approach, learning general-purpose representations from unlabeled sequence at scale. While RNA language models have demonstrated impressive performance across a broad range of downstream tasks, they generally learn from sequence reconstruction objectives alone, lacking direct connections to the regulatory principles that govern RNA function. Here we introduce Parnet, an RNA foundation model trained directly and exclusively on experimental CLIP-seq data. Parnet is a multi-task foundation model trained end-to-end on 223 eCLIP-seq experiments spanning 150 RBPs to predict base-resolution RBP binding profiles directly from RNA sequence. This CLIP-seq pretraining strategy departs fundamentally from the masked-language-modeling paradigm, anchoring learned RNA representations directly in measured protein-RNA interactions rather than sequence statistics. Parnet substantially outperforms its single-task predecessor RBPNet in binding profile and motif recovery, generalizes to unseen cell types and iCLIP data, and recapitulates position-dependent splicing regulation. Frozen Parnet embeddings, without task-specific fine-tuning, match or exceed the performance of both task-specific tools, as well as larger self-supervised RNA and genomic language models across diverse downstream tasks, including RNA biotype classification, lncRNA chromatin localization, translational efficiency, splice-site recognition, intron retention, and non-coding variant effect prediction. Importantly, Parnet remains mechanistically interpretable, tracing predictions back to the specific RBPs and motifs that drive them. These results establish the RBP interactome as a compact, functionally sufficient, and interpretable basis for foundation model pretraining in RNA biology.

bioinformatics↗

bioRxiv2026

RIBEX: Predicting and Explaining RNA Binding Across Structured and Intrinsically Disordered Regions (IDR)-rich Proteins

MotivationRNA-binding proteins (RBPs) regulate post-transcriptional processes, yet many remain undiscovered because RNA-binding activity often occurs outside canonical RNA-binding domains (RBDs), including within intrinsically disordered regions (IDRs) or through protein complexes. Computational methods can help identify novel RBPs, but approaches relying solely on sequence-derived features or ignoring the cellular interaction context are limited in capturing the complexity of RNA-binding behavior. To date, no framework rigorously integrates both sequence information and protein interaction context for RBP prediction. ResultsWe introduce RIBEX, a multimodal framework that combines protein language model (pLM) embeddings with protein interactome topology to improve RBP prediction and interpretation. Specifically, we integrate sequence representations with graph-derived positional encodings (PE) from the human STRING protein-protein interaction (PPI) network. PE are computed using Personalized PageRank, reduced with principal component analysis, and fused with pooled sequence embeddings through FiLM conditioning, while Low-Rank Adaptation (LoRA) enables parameter-efficient task adaptation. Across both an annotation-based benchmark and experimental RNA Interactome Capture (RIC) dataset, PE consistently improves predictive performance, indicating that interactome topology provides complementary information beyond sequence features. LoRA adaptation of ESM2-650M further yields larger gains than simply scaling frozen backbone size. RIBEX outperforms state-of-the-art methods such as RBP-TSTL and HydRA, particularly on challenging subsets including proteins lacking canonical RBDs and those enriched in IDRs. For interpretability, we combine sequence-level computational alanine scanning with network-level positional-encoding ablation and inverse-PCA mapping, recovering known RNA-binding domains, IDR-associated contributions, and functional interactome communities linked to RBP predictions.

bioinformatics↗

bioRxiv2025

Benchmarking Chemical, Genetic, and Cell Line Encodings for Cancer Perturbation Response Prediction

Estimating the response of tumor cells to specific perturbations is crucial for identifying effective treatments that selectively target cancer cells while sparing healthy ones, enabling personalized medicine approaches. Large-scale initiatives, such as DepMap, have profiled cancer cell line responses to various drug treatments and gene knockouts, facilitating the development of computational models that predict sensitivity of cancer cells to different perturbations. Existing models utilize diverse methods for encoding perturbations, including various chemical fingerprints and types of gene-gene relationships. They also rely on different architectures and are often trained on distinct datasets. This variability makes it unclear which chemical, genetic, or cell line encoding is most informative for predicting cancer cell viability following perturbation treatment. To address this gap, we systematically evaluated various approaches to encode chemical and genetic perturbations and cell lines on the tasks of predicting cell viability and gene dependency. We found that for genetic perturbations, STRING-based encodings yield the highest performance, considerably outperforming GO-term and protein language model based encodings, which showed promising results in previous perturbation prediction studies. For chemical perturbations, while most encoders showed comparable performance, those pre-trained on other bio-assay data yielded the highest performance. Finally, we found that for cell line encodings, raw gene expression features outperformed more sophis-ticated approaches, such as transcriptomics foundation model embeddings, as well as genotype-based encodings. Together, our results identify promising approaches for encoding chemical and genetic perturbations and enable virtual screening for perturbations with selective toxicity.

bioinformatics↗

bioRxiv2023

DeepLocRNA: An Interpretable Deep Learning Model for Predicting RNA Subcellular Localization with domain-specific transfer-learning

Accurate prediction of RNA subcellular localization plays an important role in understanding cellular processes and functions. Although post-transcriptional processes are governed by trans-acting RNA-binding proteins (RBPs) through interaction with cis-regulatory RNA motifs, current methods do not incorporate RBP-binding information. In this paper, we propose DeepLocRNA, an interpretable deep-learning model that leverages a pre-trained multi-task RBP-binding prediction model to predict the subcellular localisation of RNA molecules via fine-tuning. We constructed DeepLocRNA using a comprehensive dataset with variant RNA types and evaluated it on held-out RNA species. Our model achieved state-of-the-art performance in predicting RNA subcellular localization in mRNA and miRNA. It has demonstrated great generalization capabilities, not only for human RNA but also for mice. Moreover, the interpretability of the model is enhanced through the motif analysis, enabling the understanding of the signal factors that contribute to the predictions. The proposed model provides general and powerful prediction abilities for different RNA and species, offering valuable insights into the localisation patterns of RNA molecules and contributing to advancing our understanding of cellular processes at the molecular level.

bioinformatics↗

bioRxiv2023

A Systematic Benchmark of Machine Learning Methods for Protein-RNA Interaction Prediction

AO_SCPLOWBSTRACTC_SCPLOWRNA-binding proteins (RBPs) are central actors of RNA post-transcriptional regulation. Experiments to profile binding sites of RBPs in vivo are limited to transcripts expressed in the experimental cell type, creating the need for computational methods to infer missing binding information. While numerous machine-learning based methods have been developed for this task, their use of heterogeneous training and evaluation datasets across different sets of RBPs and CLIP-seq protocols makes a direct comparison of their performance difficult. Here, we compile a set of 37 machine learning (primarily deep learning) methods for in vivo RBP-RNA interaction prediction and systematically benchmark a subset of 11 representative methods across hundreds of CLIP-seq datasets and RBPs. Using homogenized sample pre-processing and two negative-class sample generation strategies, we evaluate methods in terms of predictive performance and assess the impact of neural network architectures and input modalities on model performance. We believe that this study will not only enable researchers to choose the optimal prediction method for their tasks at hand, but also aid method developers in developing novel, high-performing methods by introducing a standardized framework for their evaluation.

bioinformatics↗

bioRxiv2022

Towards In-Silico CLIP-seq: Predicting Protein-RNA Interaction via Sequence-to-Signal Learning

AO_SCPLOWBSTRACTC_SCPLOWUnraveling sequence determinants which drive protein-RNA interaction is crucial for studying binding mechanisms and the impact of genomic variants. While CLIP-seq allows for transcriptome-wide profiling of in vivo protein-RNA interactions, it is limited to expressed transcripts, requiring computational imputation of missing binding information. Existing classification-based methods predict binding with low resolution and depend on prior labeling of transcriptome regions for training. We present RBPNet, a novel deep learning method, which predicts CLIP crosslink count distribution from RNA sequence at single-nucleotide resolution. By training on up to a million regions, RBPNet achieves high generalization on eCLIP, iCLIP and miCLIP assays, outperforming state-of-the-art classifiers. CLIP-seq suffers from various technical biases, complicating downstream interpretation. RBPNet performs bias correction by modeling the raw signal as a mixture of the protein-specific and background signal. Through model interrogation via Integrated Gradients, RBPNet identifies predictive sub-sequences corresponding to known binding motifs and enables variant-impact scoring via in silico mutagenesis. Together, RBPNet improves inference of protein-RNA interaction, as well as mechanistic interpretation of predictions.

bioinformatics↗

bioRxiv2022

Transfer learning reveals sequence determinants of regulatory element accessibility

Dysfunction of regulatory elements through genetic variants is a central mechanism in the pathogenesis of disease. To better understand disease etiology, there is consequently a need to understand how DNA encodes regulatory activity. Deep learning methods show great promise for modeling of biomolecular data from DNA sequence but are limited to large input data for training. Here, we develop ChromTransfer, a transfer learning method that uses a pre-trained, cell-type agnostic model of open chromatin regions as a basis for fine-tuning on regulatory sequences. We demonstrate superior performances with ChromTransfer for learning cell-type specific chromatin accessibility from sequence compared to models not informed by a pre-trained model. Importantly, ChromTransfer enables fine-tuning on small input data with minimal decrease in accuracy. We show that ChromTransfer uses sequence features matching binding site sequences of key transcription factors for prediction. Together, these results demonstrate ChromTransfer as a promising tool for learning the regulatory code.

bioinformatics↗

bioRxiv2021

Computational Mapping of the Human-SARS-CoV-2 Protein-RNA Interactome

Strong evidence suggests that human human RNA-binding proteins (RBPs) are critical factors for viral infection, yet there is no feasible experimental approach to map exact binding sites of RBPs across the SARS-CoV-2 genome systematically at a large scale. We investigated the role of RBPs in the context of SARS-CoV-2 by constructing the first in silico map of human RBP / viral RNA interactions at nucleotide-resolution using two deep learning methods (pysster and DeepRiPe) trained on data from CLIP-seq experiments. We evaluated conservation of RBP binding between 6 other human pathogenic coronaviruses and identified sites of conserved and differential binding in the UTRs of SARS-CoV-1, SARS-CoV-2 and MERS. We scored the impact of variants from 11 viral strains on protein-RNA interaction, identifying a set of gain-and loss of binding events. Lastly, we linked RBPs to functional data and OMICs from other studies, and identified MBNL1, FTO and FXR2 as potential clinical biomarkers. Our results contribute towards a deeper understanding of how viruses hijack host cellular pathways and are available through a comprehensive online resource (https://sc2rbpmap.helmholtz-muenchen.de).

bioinformatics↗