LinaDongXMU
commented
2 years ago Hello Min, Thanks for your question. In the results part, we tested different combination of feature sets including AutoDock Vina(V), X-Score(X), ligand-related(L), pocket-related(P) and full protein-related(F) features. However, our results show that the combination of X, L, P and F has the best performance. Therefore, XLPFE uses X-Score(X), ligand-related(L), pocket-related(P) and full protein-related(F) features which is our final published method. E is the machine learning algrithm (extreme tree) that we use.
Best, Lina Dong ------------------ 原始邮件 ------------------ 发件人: "LinaDongXMU/XLPFE" @.>; 发送时间: 2022年8月11日(星期四) 中午1:54 @.>; @.***>; 主题: [LinaDongXMU/XLPFE] usage of VINA terms (Issue #1)
hi authors,
in the paper, https://pubs.acs.org/doi/10.1021/acsomega.2c01723
XLPFE uses 58 terms from VINA as input features.
however, in the source code, I do not see how these 58 terms are used anywhere. could you clarify please?
thank you! best, Min Htoo
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you are subscribed to this thread.Message ID: @.***>
hi authors,
in the paper, https://pubs.acs.org/doi/10.1021/acsomega.2c01723
XLPFE uses 58 terms from VINA as input features.
however, in the source code, I do not see how these 58 terms are used anywhere. could you clarify please?
thank you! best, Min Htoo