ESM-TFpredict
an explainable method for transcription factor prediction within protein sequences
In this study, we propose an ESM-TFpredict model, which leverages a pre-trained protein language model to encode amino acid sequences, followed by 1-D convolutional neural networks for transcription factor prediction.
The architecture of ESM-TFpredict model
To elucidate the model’s decision-making, we employ an integrated gradients technique to highlight the important features driving TF identification. The experimental results demonstrate that the TF-related regions have dominant influences on TF prediction task.
TF prediction attribution score of Zinc finger and BTB domain-containing protein 32 (Zinc finger position: 373-395, 401-423, and 428-450)
References
2023
- Explainable Transcription Factor Prediction with Protein Language Models (accepted)In 2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM), 2023