Results 12 comments of YRH

Sorry, I'm not familiar with Colab. I guess a mismatched version of numpy causes this error but I don't know why this happens.

> Also, is there an updated version with a more recent torch? It has been very hard to configure the environment with the old CUDA 10 to support torch 1.0.1....

The paper has a "Complexity Analysis" section and it may be helpful for you.

不是很确定为什么,BLAST输出就有一个0~1之间的identity,然后cutoff是0.6,我用的BLAST迭代次数是1

哦,是,要跑训练集的

是的,BLAST的输出结果里直接就有identity,然后是选所有hsp里最大的

应该是psiblast没跑成功,删掉psiblast生成的文件重新跑一次试试看

It's strange. I tried `dtrx` again in my Ubuntu machine and it's work well. Could you please give me more information?

OK, give me your e-mail address, but I'm not sure I can send a such large auxiliary.

困难蛋白质是用psiblast的结果得到的。 HUMAN和MOUSE是根据蛋白质的物种得到的。 困难蛋白质需要对训练集跑一下psiblast然后提取一下,go文件中蛋白质的物种信息,也需要自己提取一下HUMAN和MOUSE的数据集。 然后用这些蛋白质作为测试集就可以得到结果了。 因为是21年的代码了,所以我手上暂时没有这些测试集了。 希望可以帮助到你。