Work

A multi-head attention-based neural network for cancer-specific identification of m7G modification sites using hybrid features

Year: 2026

Type: article

Abstract: Objective N 7 -methylguanosine ( m 7 G ) is one of the most prevalent and biologically important ribonucleic acid modifications. While several computational models exist for m 7 G site prediction in human and non-human species, none have been specifically designed to detect m 7 G modification sites in cancers. This study addresses this gap by developing a cancer-specific predictive framework for m 7 G site identification. Methods We curated four m 7 G benchmark datasets from two cancer cell lines (human osteosarcoma and human hepatocellular carcinoma) from the Gene Expression Omnibus repository and included an additional benchmark dataset from m 7 G Hub v2.0. We developed a multi-head attention-based convolutional neural network model using two feature encoding methods—trinucleotide composition and one-hot encoding. The model’s performance was compared to support vector machine, random forest, extreme gradient boosting, and a convolutional neural network across all five datasets. Robustness and generalizability were evaluated using 5-fold cross-validation and independent testing. Model interpretability was examined with two explanation tools. Results The proposed model achieved superior performance across all five benchmark datasets, with marked improvements in accuracy, sensitivity, specificity, Matthews correlation coefficient, and area under the receiver operating characteristic curve compared to other machine learning and deep learning models. The interpretability analyses provided biological insight into the feature representations and decision-making process of the model. Conclusions This study presents one of the first cancer-specific computational frameworks for m 7 G site prediction. By combining multi-head attention with convolutional neural networks and complementary feature encoding, our approach enhances predictive power and interpretability. The framework offers a valuable tool for elucidating disease mechanisms, mapping epitranscriptomic signatures, and accelerating biomarker and therapeutic discovery in cancer research. The source code and curated dataset are publicly available at https://github.com/malikmtahir/m7G .

Source: Intelligent Oncology

DOI: 10.1016/j.intonc.2026.100089

Authors: Muhammad Tahir, Sheela Ramanna, Qian Liu

Institutions: Abdul Wali Khan University Mardan, The University of Winnipeg, University of Manitoba

Language: English

Topic: RNA modifications and cancer

Subfield: Molecular Biology

Field: Biochemistry, Genetics and Molecular Biology

Domain: Life Sciences

SDG: Good health and well-being

Open Access status: diamond

Funder: Natural Sciences and Engineering Research Council of Canada