LncPNdeep: A long non-coding RNA classifier based on large language model with peptide and nucleotide embedding.

LncPNdeep: A long non-coding RNA classifier based on large language model with peptide and nucleotide embedding

A concatenated deep neural network that incorporates both peptide and nucleotide embedding from masked language modeling (MLM), being able to discover complex associations between sequence information and lncRNA classification.

A Sequence Feature-based Deep Learning Framework for Automatic Family Classification of the Small Non-Coding RNA

Small non-coding RNAs (ncRNAs) are functionally diverse RNA molecules involved in various regulatory and cellular processes. Accurate family classification of ncRNAs is important for large-scale functional annotation. However, many existing computational approaches rely heavily on predicted secondary structural information, which may not always be available. In this work, we propose a purely sequence-based deep learning framework for small ncRNA family classification without using explicit secondary structure features. Our approach combines one-hot encoded sequences, sequence-level motifs, normalized length, GC content, and biologically motivated handcrafted descriptors within a DenseNet-inspired convolutional neural network architecture. The proposed model was evaluated on the benchmark nRC dataset containing 13 ncRNA families using 10fold cross-validation and an independent test set. Our method achieved an MCC of 0.951 and an F1-score of 0.954, outperforming existing sequence-based as well as several structure-aware methods. The model also demonstrated improved performance for challenging classes such as miRNA, CD-box, HACA-box, and riboswitches. In addition, we evaluated the generalization capability of our framework on the recently introduced NCY dataset, where it maintained strong and consistent classification performance. These results show that enriched sequence-derived representations can effectively capture discriminative functional signals for ncRNA family classification without relying on secondary structure prediction. The source code is available at https://github.com/sharma-nandita/small_RNA_classification.

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

Signatures of micropeptides encoded by lncRNAs in cancer progression and metastasis.

BACKGROUND Long non-coding RNAs (lncRNAs) regulate gene expression, chromatin organization, and cellular signaling. Although traditionally considered non-coding, 21% of the ~190,000 annotated lncRNA transcripts contain poorly characterized open reading frames with unknown function. METHODS We systematically identified lncRNAs encoding micropeptides (MPs) using integrated computational and experimental evidence. Expression profiles across 17 cancer types from The Cancer Genome Atlas (TCGA) were analyzed to identify cancer-associated and transitional lncRNAs (Tr-lncRNAs). Structural modeling using AlphaFold was further applied to predict folding. RESULTS We identified 478 lncRNA genes encoding 1782 MPs (10-100 amino acids). These MPs exhibit distinct amino acid and dipeptide compositions and are enriched for specific 4-mer motifs compared with canonical proteins. A subset of lncRNAs, including TNN-AS1, PVT1, XIST, and SNHG family members, encode multiple MPs. Analysis across cancer stages identified 2399 Tr-lncRNAs, most of them were cancer type and stage specific. Among these, 314 highly confident MPs from 72 Tr-lncRNAs were further analyzed. Pan-cancer analysis suggested MP-like functions for Tr-lncRNAs such as LINC01234, HAND2-AS1, XIST, UCA1, and HOXA11-AS. While most MPs are predicted to be intrinsically disordered, 3D structural modeling revealed several MPs with stable folds, including ubiquitin-like and RNase H-like structures. CONCLUSIONS Tr-lncRNA-derived MPs represent a previously underexplored class of potentially functional molecules associated with cancer clinical annotation and may serve as biomarkers for disease progression.

Scanning transcriptomes for nonlinear, domain-level similarities using hmSEEKR

Long noncoding RNAs (lncRNAs) play roles in gene regulation across kingdoms of life. However, lncRNAs with related functions often lack linear sequence similarity, making it difficult to leverage studies of one lncRNA to inform the understanding of others. We describe a k-mer-based hidden Markov model, hmSEEKR, that enables the scanning of transcriptomes for regions of non-linear sequence similarity to a query domain, without prior knowledge of where within the transcriptome the similarities may be located. When individual lncRNA domains were used as search features, hmSEEKR successfully identified regions in other RNAs that harbor non-linear sequence similarity and bind similar sets of proteins. Applying hmSEEKR to transcriptome-wide searches, we found that certain domains within the lncRNAs XIST, NEAT1, and MALAT1 exhibited widespread regional similarity to both lncRNA and protein-coding genes, while others were more unique, exhibiting similarity to ∼100 genes or fewer. Combinatorial searches uncovered RNAs containing sequential matches to core functional domains of XIST and NEAT1, and eCLIP-inferred protein-interaction networks within these RNAs more closely resembled those of XIST and NEAT1, respectively, than would be expected by chance, suggesting the searches recovered RNAs with similar biological properties. Finally, within annotated sets of cis-activating and cis-repressive lncRNAs, we observed opposing enrichments for similarity to domains associated with transcription-promoting complexes and heterogeneous nuclear ribonucleoprotein (hnRNP) binding, respectively, suggesting the enriched sequences may contribute to regulatory functions. hmSEEKR can be applied with minimal training data and enables the a priori discovery of RNA domains that share nonlinear similarity, offering a sequence-informed approach to discover functional elements within noncoding transcriptomes.

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

Protein methylation is a crucial post-translational modification (PTM) responsible for many diseases and accurate prediction of the methylation site is important for understanding the molecular mechanism of the disease. The models have been successful in capturing contextual dependencies in protein sequences, with deep learning models, specifically those based on the Transformer architecture and Multi-Head Attention, exhibiting good performance. However, most existing techniques rely on the sequence-only or hand-crafted features and are unable to capture biochemical properties and positional patterns, thereby limiting cross-species generalization and prediction accuracy. To cater for such demands, PLM-ArgMe is presented that is based on a symmetry-sensitive Transformer framework using context-aware ESM-2 residue embeddings, which is mapped through a novel Bio-Symmetric Mirrored Sinusoidal Encoding (BSMSE) strategy to address the biological symmetry hypothesis of arginine methylation. ESM-2 encodes evolutionary and structural context, while biochemical representations are enhanced by physicochemical features. A symmetry-aware positional encoding strategy and bidirectional multi-head self-attention are used to model structural, sequence-level, and feature-level dependencies. The proposed framework, PLM-ArgMe, achieves prediction accuracies of 90.91%, 93%, 87.44%, and 87.22% on Chimpanzee, Rat, Human, and Mouse datasets, respectively. When trained and evaluated on a combined multi-species dataset, the model attains an overall accuracy of 88.41%. The results reveal good generalization on a variety of datasets and suggest that PLM-ArgMe is a robust method for arginine methylation site prediction.