PLM-ArgMe: Protein language model for arginine methylation prediction for different species.

ESM2-Kcr: a novel model combining protein language model and deep learning methods for predicting lysine crotonylation sites

DNCLA: A Deep Learning Model for TFBS Identification Based on Structural and Conformational Properties of Nucleotides and Dinucleotides

Identifying transcription factor binding sites (TFBSs) is fundamental to understanding complex gene regulatory mechanisms and the functions of non-coding regions. Although existing methods have achieved substantial strides, capturing both local structural features and long-range spatial dependencies within DNA sequences remains a major challenge for improving prediction accuracy. In this study, we propose DNCLA, a deep learning model that synergizes multisize convolutional fusion, Bidirectional Long ShortTerm Memory (Bi-LSTM) networks, and a multi-head self-attention mechanism. At the feature extraction level, DNCLA breaks through the limitations of traditional single-sequence encoding by fusing Nucleotide Chemical Properties (NCP) with Dinucleotide Physicochemical Properties (DPCP). NCP provides a refined characterization of chemical differences between bases based on ring structures, hydrogen bond sites, and functional group properties, while DPCP introduces parameters such as local structural stability and geometric flexibility of the DNA. Subsequently, the model extracts spatial evolution from these high-dimensional features through a multi-size convolutional module; captures long-range spatial dependencies using Bi-LSTM layers; and employs a multi-head self-attention mechanism to achieve adaptive weight distribution of global features, thereby enhancing the perception of key regulatory motifs. Results from training and testing the proposed model on 165 ChIPseq datasets demonstrate that DNCLA possesses robust generalization capabilities and high predictive performance in TFBSs identification. This suggests that the incorporation of physicochemical features better elucidates the essence of interactions between transcription factors and DNA.

Structure-aware deep learning enhances m6A prediction and reveals cell type-associated RNA structural signatures

N6-methyladenosine (m6A), the most abundant mRNA modification in eukaryotes, plays essential roles in gene regulation and disease pathogenesis. Computational prediction of m6A sites offers a scalable alternative to costly experimental approaches, yet current methods rely predominantly on linear sequence features. This overlooks potentially informative RNA structural context, which is associated with local methylation patterns and may provide complementary predictive information beyond linear sequence motifs. To incorporate this complementary information, we propose SMART-m6A (Sequence–structure Multifeature Attention RNA Transformer for m6A), a deep learning framework that integrates sequence and structural information through parallel convolutional feature extraction and structure-guided attention for multifeature fusion. SMART-m6A achieves superior predictive performance compared to existing methods, with particularly clear advantages in sequence-ambiguous candidates. Beyond prediction accuracy, learned attention patterns reveal strong concordance with experimentally validated m6A-binding protein recognition sites and identify potentially novel regulatory motifs. Through systematic ablation studies and targeted structural-input perturbation analyses, we show that sequence and structure provide complementary predictive information, and that sites with greater prediction sensitivity to structural perturbation exhibit distinct local structural profiles between cell lines. Collectively, this work demonstrates the predictive value of sequence-derived structural features in m6A modeling and provides a multifeature deep learning framework for accurate and interpretable structure-aware epitranscriptomic prediction.

Learning Shared Residue Backgrounds and Modification-Specific Offsets for PTM Site Prediction

Post-translational modifications (PTMs) are chemical changes added to proteins after translation. These changes affect protein function and regulation, and their disruption is linked to disease-associated mechanisms. Because experimentally validating all possible modification sites is impractical, many computational predictors have been developed for PTM site prediction. In this work, we study whether a shared model can represent common residue-background patterns while learning modification-specific background-to-positive offsets. This framing is especially relevant for residues such as lysine (K), which can be acetylated, ubiquitinated, methylated, or sumoylated depending on the surrounding protein context. We propose an anchor-guided rectified flow matching framework for multi-type PTM site prediction from protein language model embeddings. For each PTM–residue pair, the model builds residue-background anchors from PTM-compatible unannotated residues and positive anchors from experimentally annotated modified residues. Given a candidate residue and target modification type, the model compares the residue embedding with these anchor sets and uses a rectified flow module to estimate a modification-conditioned background-to-positive offset. This offset is combined with anchor-based features and used for site scoring. We evaluate the framework on a dbPTM-derived benchmark covering six commonly studied PTMs: phosphorylation, acetylation, ubiquitination, methylation, sumoylation, and N-linked glycosylation. In the shared-model setting, our approach achieves a macro AUPRC of 0.4195, improving over the gated multi-anchor baseline of 0.4154, while independently trained per-modification models achieve 0.4353. These results suggest that multi-type PTM prediction can be modeled within a single shared framework by combining residue-background anchors with modification-conditioned offset features.

ProtSyntax: a protein large language model for decoding post-translational modification syntax and function

Post-translational modifications (PTMs) expand protein function by encoding context-dependent regulatory states, and their dysregulation contributes to cancer, neurodegeneration and metabolic disease. However, existing methods treat PTMs as independent residue labels, limiting their ability to distinguish contextually permissible sites, model crosstalk and infer functional consequences. Here we introduce ProtSyntax, a PTM-aware foundation protein language model combining protein-aware positional encoding, bidirectional state-space propagation, geometry-constrained attention and adaptive multi-objective learning. This design integrates residue chemistry, motif order, long-range context and three-dimensional microenvironments while coupling PTM recognition to enzyme function. Across 40 PTM-site benchmarks, ProtSyntax exceeded the strongest baselines in mean MCC and AP by 12.66% and 10.67%. ProtSyntax also recovered masked PTM types and sites, rejected structural decoys, generalized to data-scarce modifications, reconstructed crosstalk and linked PTM perturbations to enzyme kinetics. Applications to pathogenic variants, biomolecular condensates and disease-associated PTM landscapes demonstrate its potential to decode the regulatory language of the modified proteome.

Identifying RNA ac4C Modification Sites via Pseudo-Nucleotide Fingerprint Encoding and Multi-Scale Feature Integration.

DFM-ac4C, a novel computational framework designed for the accurate prediction of ac4C modification sites, significantly outperforms existing models, achieving outstanding predictive metrics, and underscores DFM-ac4C's effectiveness as a robust and efficient tool for RNA ac4C site identification.