Department of Mathematics and Informatics, Freie Universität Berlin, Germany.
The Department of Molecular Biosciences, The Wenner-Gren Institute, Stockholm University, Stockholm, Sweden.
PLoS Comput Biol. 2024 Jul 25;20(7):e1012253. doi: 10.1371/journal.pcbi.1012253. eCollection 2024 Jul.
Structure prediction of protein complexes has improved significantly with AlphaFold2 and AlphaFold-multimer (AFM), but only 60% of dimers are accurately predicted. Here, we learn a bias to the MSA representation that improves the predictions by performing gradient descent through the AFM network. We demonstrate the performance on seven difficult targets from CASP15 and increase the average MMscore to 0.76 compared to 0.63 with AFM. We evaluate the procedure on 487 protein complexes where AFM fails and obtain an increased success rate (MMscore>0.75) of 33% on these difficult targets. Our protocol, AFProfile, provides a way to direct predictions towards a defined target function guided by the MSA. We expect gradient descent over the MSA to be useful for different tasks.
蛋白质复合物的结构预测随着 AlphaFold2 和 AlphaFold-multimer (AFM) 的出现有了显著的提高，但只有 60%的二聚体能够被准确预测。在这里，我们通过在 AFM 网络中进行梯度下降，学习了一种对 MSAs 表示的偏差，从而提高了预测的准确性。我们在 CASP15 的七个困难目标上展示了该方法的性能，与 AFM 相比，平均 MMscore 提高到了 0.76。我们在 487 个 AFM 失败的蛋白质复合物上评估了该方法，在这些困难目标上，成功率（MMscore>0.75）提高到了 33%。我们的方法 AFProfile 提供了一种通过 MSAs 引导，将预测定向到特定目标函数的方法。我们期望通过 MSAs 进行梯度下降对于不同的任务都是有用的。
