# 多肽和蛋白的化学位移计算 ** 这个项目的主要目的是辅助核磁谱图归属,特别是直接从系列可以预测结构,然后结构预测化学位移,归属HSQC谱图的指认 ** 1. 以下这篇文章是说用QM来计算生物大分子的化学位移,Using quantum chemistry to estimate chemical shifts in biomolecules, 他们还开发一个[AFNMR](https://github.com/dacase/afnmr)软件来实现。我的想法是通过这个来计算多肽的化学位移,来辅助多肽的核磁谱图指认。当然如果计算运行的情况下,也是可以试试蛋白的计算。 2. 接下来一片比较新的预测蛋白化学位移的文献 Accurate prediction of chemical shifts for aqueous protein structure on “Real World” data。我没有搭建起来,可以试试。也提供的源码,https://github.com/THGLab/CSpred 目前好像是PDB格式问题,不知道,就是不给出结果。还得试试。 当然还有其他的预测软件,比如SPARTA+https://spin.niddk.nih.gov/bax/software/SPARTA+/, SHIFTX2, PROSHIFT,CamShift。 # 小分子化学位移计算 ### 构象搜索 SDEGen: Learning to Evolve Molecular Conformations from Thermodynamic Noise for Conformation Generation https://pubs.rsc.org/en/content/articlelanding/2023/sc/d2sc04429c 采用随机微分方程模拟分子构象,联合概率深度学习的DDIM模型,提高效率和精度。可以用该模型和科音的molclus模型,还有XTB模型对比。我这边主要是要用来计算核磁化学位移的。https://github.com/HaotianZhangAI4Science/SDEGen 该文章提供源代码,可以后面试着用用。 - XTBDFT: Automated workflow for conformer searching of minima and transition states powered by extended tight binding and density functional theory https://github.com/sibo/xtbdft - CREST https://github.com/grimme-lab/crest - molclus http://www.keinsci.com/research/molclus.html ### 自动化计算核磁共振化学位移 - An automated framework for high-throughput predictions of NMR chemical shifts within liquid solutions 小分子的化学位移预测 - Elucidating Structures of Complex Organic Compounds Using a Machine Learning Model Based on the 13C NMR Chemical Shifts 这一篇也不错,https://github.com/fenglb/SVM-M - AI预测化学位移的方法:Transfer Learning from Simulation to Experimental Data: NMR Chemical Shift Predictions, CASCADE,Real-time prediction of 1H and 13C chemical shifts with DFT accuracy using a 3D graph neural network ### 量化自动化计算核磁共振全谱 https://github.com/grimme-lab/enso Fully Automated Quantum-Chemistry-Based Computation of Spin–Spin-Coupled Nuclear Magnetic Resonance Spectra ![](https://onlinelibrary.wiley.com/cms/asset/e2011bb9-a812-412f-8be0-d32635c47da8/anie201708266-fig-0006-m.png)