Think about adding `templating` when generating TS guesses
还没有人认领这个 Issue。
评估
调研方向
No implementation file, test, or entry point is named. Start by locating the code that generates transition-state guesses and compare its current approach with the templating workflow described in the issue and paper. Done means the scope and design are established, including template matching and atom correspondence.
由索引模型根据 Issue 内容生成。
描述
From reading about other tools, many will template when generating a TS guess. From the AutoTS paper
In addition to interpolation, we have the ability to use geometric information from a previously located TS. We refer to this approach as templating and to the predefined TS geometries as templates. TS templates, along with the reactant and product geometries that they connect (as verified by an IRC calculation), are stored in an auxiliary computer file. Several string representations of the input reaction are generated and used to search this template file for matching reactions. Each representation is of a particular order, i, and is constructed by generating SMILES strings for the atoms in the reaction center plus all (i – 1)th nearest neighbors. We tend to use only three orders starting with i = 1. The SMILES patterns of the reactant and product structures are combined with the symbol “≫” as in reaction SMILES. The algorithm has the preference for matches at the highest order attainable.
Once it is recognized that a usable template exists, we extract it from the file and compute an atom correspondence between the input structures and template structures. This is done by utilizing an MCS algorithm, which computes a bijection for the reaction center atoms. The reaction matching described above guarantees that the reaction center and possibly some number of neighbors make up a substructure shared between the reactant and product.
- 主要语言
- Python
- 星标
- 51
- 派生
- 19
- PR 合并指标
- 30 天内没有已合并 PR
环境准备
这个项目没有提供开发容器、Dockerfile 或贡献指南，环境需要你自己搭建：先看它的 README，通用步骤见我们的新手贡献指南。
从这里开始
- 先读完整个 Issue，再读项目的贡献指南。
- 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
- Fork 仓库，在一个分支上完成修改。
- 提交 Pull Request，并在描述里引用这个 Issue 编号。
ReactionMechanismGenerator/AutoTST 的其他 Issue
-
Conformer generation isn't deterministic可能重新可做 @sevyharris 于 614 天前认领，目前没有进行中的 PR。 未关闭
ReactionMechanismGenerator/AutoTST#94 · 2 条评论 · 已指派 1 人 ·
-
难度 4/5 3-5 天 新手友好度 30/100
ReactionMechanismGenerator/AutoTST#91 · 2 条评论 ·
-
难度 3/5 1-2 天 新手友好度 25/100
ReactionMechanismGenerator/AutoTST#81 · 3 条评论 ·
-
log to TS未关闭
难度 5/5 一周以上 新手友好度 25/100
-
难度 4/5 3-5 天 新手友好度 35/100
ReactionMechanismGenerator/AutoTST#73 · 1 条评论 ·
查看 ReactionMechanismGenerator/AutoTST 的全部 Issue
相似的 Issue
-
request-theme
难度 1/5 1 小时以内 新手友好度 66/100
LizardByte/ThemerrDB#9575 ·
维护者通常 1 天内回复
-
author-page correction
难度 2/5 1-3 小时 新手友好度 62/100
acl-org/acl-anthology#10599 · 1 条评论 ·
维护者通常 1 天内回复
-
难度 2/5 1-3 小时 新手友好度 78/100
zilliztech/memsearch#818 ·
维护者通常 1 天内回复
-
[FAQ] Commit ID for Projects可能已有人在做 @github-actions 今天认领。 未关闭faq-proposal
难度 1/5 1 小时以内 新手友好度 84/100
DataTalksClub/faq#503 · 1 条评论 ·
维护者通常 2 天内回复
-
README.zh-CN/ja/es still hard-code 501 test counts after README.md went count-free可能已有人在做 @Aaqibhafeezkhan 今天认领。 未关闭beginner documentation good first issue
难度 2/5 1-3 小时 新手友好度 75/100
BistaDinesh03/critique#164 · 1 条评论 ·
维护者通常 1 天内回复