psp3dcg
psp3dcg
> Hi I am new to codes but I would like to use Alphafold to predict some of my proteins. I met the same problem. I used the same software...
I usually use MSA generated by hhblits as input for hhsearch.
OK, thank you for the reply~
> Hi @psp3dcg, I saw you reopened this. Anything else I can do here? Oh, yes. I want to ask the name of protein families you used during the training...
> 他的教程确实写得不太清楚,我跑通了一遍大概是:输入是tsv格式的序列文件,用do_embedding.py先将其转换为embedding,然后用do_agg.py将上一步输出整合为index文件,前两步完成后,可以用do_retrieval.py进行序列查找。但uniref数据库的embedding非常缓慢。。一张v10032g预计得跑300+h,我直接放弃了。。 你好,我看UniRef90.fasta中一条序列都是30000+氨基酸,这种长序列怎么做embedding提取啊?