Dataset for Discovery of Self-Assembling π-Conjugated Peptides by Active Learning-Directed Coarse-Grained Molecular Simulation

This dataset contains 186 molecular dynamics simulation trajectories of coarse-grained pi-conjugated peptides in the Asp-X-X-X-(oligophenylenevinylene)3-X-X-X-Asp (DXXX-OPV3-XXXD) family, where X denotes one of the 20 natural amino acids. The oligopeptide wings are constrained to be mirror-symmetric both in the identity of the amino acids and the N-to-C directionality such that each molecule possesses two C-termini. These trajectories were generated during an active learning search for the peptide sequences that yield structurally optimal self-assembled pseudo-1D nanoaggregates with good stacking between the pi-cores. Directories containing the trajectories are named "dxxx" where each "x" corresponds to an amino acid in the DXXX-OPV3-XXXD sequence (e.g. 'dfag' corresponds to the structure DFAG-OPV3-GAFD). Each trajectory itself is a 3,000 nanosecond simulation of 96 identical pi-conjugated peptides in water (/solvent), and the same trajectories where the water has been removed (/no-solvent).

本数据集包含186条粗粒化π共轭肽的分子动力学模拟轨迹，所属家族为天冬氨酸（Asp）-X-X-X-(低聚亚苯基亚乙烯基（oligophenylenevinylene）)3-X-X-X-天冬氨酸（Asp）（DXXX-OPV3-XXXD），其中X代表20种天然氨基酸之一。该寡肽侧翼在氨基酸种类及N到C的方向上均受镜像对称约束，因此每个分子均拥有两个C端。这些轨迹生成于一项主动学习搜索流程中，旨在筛选可形成结构最优、且π核间堆叠效果优异的自组装准一维纳米聚集体的肽序列。存储轨迹的目录命名为“dxxx”，其中每个“x”对应DXXX-OPV3-XXXD序列中的一个氨基酸位点，例如“dfag”对应结构DFAG-OPV3-GAFD。每条轨迹均为在水（/溶剂）环境下对96个相同π共轭肽开展的3000纳秒模拟，同时附带移除水分子后的同轨迹副本（/no-solvent）。