Help
About UniDoc
The input to UniDoc is protein sequence or 3D structure, as shown in Figure 1, the UniDoc works as follows.
(1) When protein structure is submitted, the distance matrix is extracted from 3D structure. Then, we use the hierarchical clustering to decompose the protein.
(2) When protein sequence is submitted, the distance matrix predicted by our recent deep learning based structure prediction algorithm trRosetta. Then, Then, we use the hierarchical clustering to decompose the protein.
Figure 1. The flowchart of the UniDoc algorithm.
Structure-based domain parsing result
Figure 2. The structure-based domain parsing result.
Sequence-based domain recognition result
Figure 3. The sequence-based domain recognition result.
Need more help?
If you have more questions or comments about the server, please email yangjy
sdu.edu.cn.