Prediction of viral oncoproteins through the combination of generative adversarial networks and machine learning techniques DOI Creative Commons
Jorge F. Beltrán,

Lisandra Herrera-Belén,

Alejandro J. Yáñez

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Ноя. 7, 2024

Viral oncoproteins play crucial roles in transforming normal cells into cancer cells, representing a significant factor the etiology of various cancers. Traditionally, identifying these is both time-consuming and costly. With advancements computational biology, bioinformatics tools based on machine learning have emerged as effective methods for predicting biological activities. Here, first time, we propose an innovative approach that combines Generative Adversarial Networks (GANs) with supervised to enhance accuracy generalizability viral oncoprotein prediction. Our methodology evaluated multiple models, including Random Forest, Multilayer Perceptron, Light Gradient Boosting Machine, eXtreme Boosting, Support Vector Machine. In ten-fold cross-validation our training dataset, GAN-enhanced Forest model demonstrated superior performance metrics: 0.976 accuracy, F1 score, 0.977 precision, sensitivity, 1.0 AUC. During independent testing, this achieved 0.982 These results establish new tool, VirOncoTarget, accessible via web application. We anticipate VirOncoTarget will be valuable resource researchers, enabling rapid reliable prediction advancing understanding their role biology.

Язык: Английский

ToxinPred 3.0: An improved method for predicting the toxicity of peptides DOI
Anand Singh Rathore, Shubham Choudhury, Akanksha Arora

и другие.

Computers in Biology and Medicine, Год журнала: 2024, Номер 179, С. 108926 - 108926

Опубликована: Июль 21, 2024

Язык: Английский

Процитировано

48
Leveraging transformers‐based language models in proteome bioinformatics DOI
Nguyen Quoc Khanh Le

PROTEOMICS, Год журнала: 2023, Номер 23(23-24)

Опубликована: Июнь 29, 2023

Abstract In recent years, the rapid growth of biological data has increased interest in using bioinformatics to analyze and interpret this data. Proteomics, which studies structure, function, interactions proteins, is a crucial area bioinformatics. Using natural language processing (NLP) techniques proteomics an emerging field that combines machine learning text mining Recently, transformer‐based NLP models have gained significant attention for their ability process variable‐length input sequences parallel, self‐attention mechanisms capture long‐range dependencies. review paper, we discuss advancements proteome examine advantages, limitations, potential applications improve accuracy efficiency various tasks. Additionally, highlight challenges future directions these research. Overall, provides valuable insights into revolutionize

Язык: Английский

Процитировано

37
ToxinPred 3.0: An improved method for predicting the toxicity of peptides DOI Open Access
Anand Singh Rathore, Akanksha Arora, Shubham Choudhury

и другие.

bioRxiv (Cold Spring Harbor Laboratory), Год журнала: 2023, Номер unknown

Опубликована: Авг. 14, 2023

Abstract Toxicity emerges as a prominent challenge in the design of therapeutic peptides, causing failure numerous peptides during clinical trials. In 2013, our group developed ToxinPred, computational method that has been extensively adopted by scientific community for predicting peptide toxicity. this paper, we propose refined variant ToxinPred showcases improved reliability and accuracy Initially, used BLAST alignment-based toxicity prediction, yet coverage was limited. We motif-based approach with MERCI software to identify unique toxic patterns. Despite specificity gains, sensitivity compromised. alignment-free methods using machine/deep learning, achieving balance prediction. A deep learning model (ANN – LSTM fixed sequence length) one-hot encoding attained 0.93 AUROC 0.71 MCC on independent data. The machine (extra tree) compositional features achieved 0.95 0.78 MCC. Lastly, hybrid or ensemble combining two more models enhance performance. Hybrid approaches, including 0.98 0.81 Evaluation data demonstrated method’s superiority. To cater needs community, have standalone software, pip package web-based server ToxinPred3 ( https://github.com/raghavagps/toxinpred3 https://webs.iiitd.edu.in/raghava/toxinpred3/ ) . Author’s Biography Anand Singh Rathore is currently pursuing Ph.D. Computational Biology at Department Biology, Indraprastha Institute Information Technology, New Delhi, India. Akanksha Arora Shubham Choudhury Purava Tijare Project Fellow Gajendra P. S. Raghava working Professor Head Highlights Implementation alignment similarly based techniques peptides. Discovery toxicity-associated patterns identification regions Development learning-based Ensemble combine methods. Web screening peptides/proteins.

Язык: Английский

Процитировано

24
Reconstructing 3D chromosome structures from single-cell Hi-C data with SO(3)-equivariant graph neural networks DOI Creative Commons
Yanli Wang, Jianlin Cheng

NAR Genomics and Bioinformatics, Год журнала: 2025, Номер 7(1)

Опубликована: Янв. 7, 2025

Abstract The spatial conformation of chromosomes and genomes single cells is relevant to cellular function useful for elucidating the mechanism underlying gene expression genome methylation. chromosomal contacts (i.e. regions in proximity) entailing three-dimensional (3D) structure a cell can be obtained by single-cell chromosome capture techniques, such as Hi-C (ScHi-C). However, due sparsity ScHi-C data, it still challenging traditional 3D optimization methods reconstruct structures from data. Here, we present machine learning-based method based on novel SO(3)-equivariant graph neural network (HiCEGNN) HiCEGNN consistently outperforms both only other deep learning across diverse cells, different structural resolutions, noise levels Moreover, robust against

Язык: Английский

Процитировано

1
Comprehensive review of Transformer‐based models in neuroscience, neurology, and psychiatry DOI Creative Commons
Shan Cong, Hang Wang, Yang Zhou

и другие.

Brain‐X, Год журнала: 2024, Номер 2(2)

Опубликована: Апрель 26, 2024

Abstract This comprehensive review aims to clarify the growing impact of Transformer‐based models in fields neuroscience, neurology, and psychiatry. Originally developed as a solution for analyzing sequential data, Transformer architecture has evolved effectively capture complex spatiotemporal relationships long‐range dependencies that are common biomedical data. Its adaptability effectiveness deciphering intricate patterns within medical studies have established it key tool advancing our understanding neural functions disorders, representing significant departure from traditional computational methods. The begins by introducing structure principles architectures. It then explores their applicability, ranging disease diagnosis prognosis evaluation cognitive processes decoding. specific design modifications tailored these applications subsequent on performance also discussed. We conclude providing assessment recent advancements, prevailing challenges, future directions, highlighting shift neuroscientific research clinical practice towards an artificial intelligence‐centric paradigm, particularly given prominence most successful large pre‐trained models. serves informative reference researchers, clinicians, professionals who interested harnessing transformative potential

Язык: Английский

Процитировано

7
Using a hybrid neural network architecture for DNA sequence representation: A study on N4-methylcytosine sites DOI

Van‐Nui Nguyen,

Trang-Thi Ho, Thu-Dung Doan

и другие.

Computers in Biology and Medicine, Год журнала: 2024, Номер 178, С. 108664 - 108664

Опубликована: Май 27, 2024

Язык: Английский

Процитировано

6
A Computational Predictor for Accurate Identification of Tumor Homing Peptides by Integrating Sequential and Deep BiLSTM Features DOI

Roha Arif,

Sameera Kanwal,

Saeed Ahmed

и другие.

Interdisciplinary Sciences Computational Life Sciences, Год журнала: 2024, Номер 16(2), С. 503 - 518

Опубликована: Май 11, 2024

Язык: Английский

Процитировано

4
L-SSHNN: A Larger search space of Semi-Supervised Hybrid NAS Network for echocardiography segmentation DOI
Renqi Chen,

Fan Nian,

Yuhui Cen

и другие.

Expert Systems with Applications, Год журнала: 2025, Номер unknown, С. 127084 - 127084

Опубликована: Март 1, 2025

Язык: Английский

Процитировано

0
Biological Prior Knowledge-Embedded Deep Neural Network for Plant Genomic Prediction DOI Open Access
Chen Ye, Kai Li, W. Y. Sun

и другие.

Genes, Год журнала: 2025, Номер 16(4), С. 411 - 411

Опубликована: Март 31, 2025

Background/Objectives: Genomic prediction is a powerful approach that predicts phenotypic traits from genotypic information, enabling the acceleration of trait improvement in plant breeding. Traditional genomic methods have primarily relied on linear mixed models, such as Best Linear Unbiased Prediction (GBLUP), and conventional machine learning like Support Vector Regression (SVR). are limited handling high-dimensional data nonlinear relationships. Thus, deep also been applied to recent years. Methods: We proposed iADEP, Integrated Additive, Dominant, Epistatic model based learning. Specifically, single nucleotide polymorphism (SNP) integrating latent genetic interactions genome-wide association study results biological prior knowledge fused an SNP embedding block, which then input local encoder. The encoder with omic-data-incorporated global decoder through multi-head attention mechanism, followed by multilayer perceptrons. Results: Firstly, we demonstrated experiments four datasets iADEP outperforms existing genotype-to-phenotype prediction. Secondly, validated effectiveness ablation experiments. Third, provided available module for combining other omics propose novel method fusing them. Fourthly, explored impact feature selection performance conclude utilizing full set SNPs generally provides optimal results. Finally, altering partition training testing sets, investigated differences between transductive inductive Conclusions: new AI breeding, promising integrates enables combination data.

Язык: Английский

Процитировано

0
DMGAT: predicting ncRNA-drug resistance associations based on diffusion map and heterogeneous graph attention network DOI Creative Commons
Tingyu Liu,

Qiuhao Chen,

R.P. Liu

и другие.

Briefings in Bioinformatics, Год журнала: 2025, Номер 26(2)

Опубликована: Март 1, 2025

Abstract Non-coding RNAs (ncRNAs) play crucial roles in drug resistance and sensitivity, making them important biomarkers therapeutic targets. However, predicting ncRNA-drug associations is challenging due to issues such as dataset imbalance sparsity, limiting the identification of robust biomarkers. Existing models often fall short capturing local global sequence information, reliability predictions. This study introduces DMGAT (diffusion map heterogeneous graph attention network), a novel deep learning model designed predict associations. integrates diffusion maps for embedding, convolutional networks feature extraction, GAT information fusion. To address imbalance, incorporates sensitivity employs random forest classifier select reliable negative samples. embeds ncRNA sequences SMILES using word2vec technique, information. The constructs network by combining similarity Gaussian Interaction Profile kernel similarity, providing comprehensive representation interactions. Evaluated through five-fold cross-validation on curated from NoncoRNA ncDR, outperforms seven state-of-the-art methods, achieving highest area under receiver operating characteristic curve (0.8964), precision-recall (0.8984), recall (0.9576), F1-score (0.8285). raw data are released Zenodo with identifier 13929676. source code available at https://github.com/liutingyu0616/DMGAT/tree/main.

Язык: Английский

Процитировано

0