School of Information and Communication Technology, Griffith University, Queensland, Australia.
Institute of Integrated and Intelligent Systems, Griffith University, Queensland, Australia; School of Information and Physical Sciences, The University of Newcastle, New South Wales, Australia.
Comput Biol Med. 2022 Sep;148:105824. doi: 10.1016/j.compbiomed.2022.105824. Epub 2022 Jul 11.
Predicted inter-residue distances are a key behind recent success in high quality protein structure prediction (PSP). However, prediction of both short and long distance values together is challenging. Consequently, predicted short distances are mostly used by existing PSP methods. In this paper, we use a stacked meta-ensemble method to combine deep learning models trained for different ranges of real-valued distances. On five benchmark sets of proteins, our proposed inter-residue distance prediction method improves mean Local Distance Different Test (LDDT) scores at least by 5% over existing such methods. Moreover, using a real-valued distance based conformational search algorithm, we also show that predicted long distances help obtain significantly better protein conformations than when only predicted short distances are used. Our method is named meta-ensemble for distance prediction (MDP) and its program is available from https://gitlab.com/mahnewton/mdp.
预测残基间距离是近期高质量蛋白质结构预测 (PSP) 取得成功的关键。然而，预测短距离和长距离值是具有挑战性的。因此，现有的 PSP 方法大多使用预测的短距离。在本文中，我们使用堆叠元集成方法来组合针对不同实值距离范围训练的深度学习模型。在五个蛋白质基准集上，我们提出的残基间距离预测方法至少将现有方法的平均局部距离差异测试 (LDDT) 得分提高了 5%。此外，使用基于实值距离的构象搜索算法，我们还表明，预测长距离有助于获得比仅使用预测短距离更好的蛋白质构象。我们的方法名为用于距离预测的元集成 (MDP)，其程序可从 https://gitlab.com/mahnewton/mdp 获得。
