Transformer Neural Networks for Protein Family and Interaction Prediction Tasks
Transformer Neural Networks for Protein Family and Interaction Prediction Tasks
复制标题
DOI：
10.1089/cmb.2022.0132
复制
发表时间：
2022-08
期刊：
影响因子：
--
通讯作者：
Ananthan Nambiar;Simon Liu;Maeve Heflin;John Malcolm Forsyth;S. Maslov;Mark Hopkins;Anna M. Ritz
中科院分区：
文献类型：
--
作者：
Ananthan Nambiar;Simon Liu;Maeve Heflin;John Malcolm Forsyth;S. Maslov;Mark Hopkins;Anna M. Ritz
科学界正在快速生成蛋白质序列信息，但这些蛋白质中只有一小部分可以通过实验进行表征。虽然用于蛋白质预测任务的有前途的深度学习方法已经出现，但它们具有计算限制或旨在解决特定任务。我们提出了一个 Transformer 神经网络，可以预训练与任务无关的序列表示。该模型经过微调可以解决两种不同的蛋白质预测任务：蛋白质家族分类和蛋白质相互作用预测。我们的方法可与现有最先进的蛋白质家族分类方法相媲美，同时比其他架构更通用。此外，对于我们生成的三种不同场景中的两种，我们的方法优于其他蛋白质相互作用预测方法。这些结果为微调其他蛋白质预测任务的预训练序列表示提供了一个有前景的框架。
The scientific community is rapidly generating protein sequence information, but only a fraction of these proteins can be experimentally characterized. While promising deep learning approaches for protein prediction tasks have emerged, they have computational limitations or are designed to solve a specific task. We present a Transformer neural network that pre-trains task-agnostic sequence representations. This model is fine-tuned to solve two different protein prediction tasks: protein family classification and protein interaction prediction. Our method is comparable to existing state-of-the-art approaches for protein family classification while being much more general than other architectures. Further, our method outperforms other approaches for protein interaction prediction for two out of three different scenarios that we generated. These results offer a promising framework for fine-tuning the pre-trained sequence representations for other protein prediction tasks.