PaRoutes

该数据集名为PaRoutes，用于评估DirectMultiStep模型在为各种化合物生成多步合成路线方面的性能。数据集包含了用于训练和测试模型的路线路径，其中包括未包含在训练集中的经FDA批准的药物。该任务的目标是进行逆合成预测。

This dataset, named PaRoutes, is developed to evaluate the performance of the DirectMultiStep model when generating multi-step synthetic routes for diverse compounds. The dataset contains route paths for model training and testing, including FDA-approved drugs that are excluded from the training set. The objective of this task is retrosynthetic prediction.

搜集汇总

背景与挑战

背景概述

PaRoutes是一个用于基准测试多步逆合成预测的框架，包含反应数据集、路线集合和库存分子集合，支持路线质量和多样性分析。