Exploring the Potential of Deep Learning in Protein Remote Homology Detection and Folds Identification Using Transfer Learning and Attention Mechanism

@article{1705014,
      author = {K. Gopinath, G. Rajendran},
      title = {Exploring the Potential of Deep Learning in Protein Remote Homology Detection and Folds Identification Using Transfer Learning and Attention Mechanism},
      journal = {Iconic Research And Engineering Journals},
      year = {2023},
      volume = {7},
      number = {3},
      pages = {16-23},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1705014.pdf},
      abstract = {Protein remote homology detection and fold identification are important tasks in computational biology that have significant implications for understanding protein function, evolution, and drug design. Deep learning has emerged as a powerful approach for solving various biological problems, including protein remote homology detection and fold identification. In this work, the potential of deep learning in improving the accuracy of protein remote homology detection and fold identification is explored. The performance of deep learning models has been compared with that of traditional methods using the Matthews Correlation Coefficient as the evaluation metric. Our results show that deep learning models outperform traditional methods in detecting protein remote homology and identifying folds. This paper provides evidence of the potential of deep learning in improving the accuracy of protein remote homology detection and folds identification.},
      keywords = {Proteins, remote homology detection, fold identification, deep learning, Matthews Correlation Coefficient, convolutional neural networks, recurrent neural networks, transfer learning},
      month = {September},
  }