About the server

The server consists of two related modules: fast search of structure database and multiple protein structure alignment. The structure search could be performed on different databases include PDB, SCOPe, CATH, BFVD, and AlphaFoldDB. The multiple protein structure alignment is performed using the fast and accurate algorithm mTM-align.

The database search consists of two steps:

Computing the cosine similarity of the query's feature with the targets' features. Targets with high cosine similarity are considered to be the candidates for high-speed retrieval.

For high-accuracy retrieval, the candidates from high-speed retrieval are further filtered using fTM-align. Hits with TM-score higher than 0.4 are returned.

The multiple structure alignment is built with three steps:

Generating all the Pairwise structure alignments for the input structures with TM-align.

A structure-based phylogenetic tree is constructed using the UPGMA algorithm.

Progressive construction of a multiple protein structure alignment using the branching order from the phylogenetic tree.

Figure 1. The flowchart of mTM-align server.

Navigation bar

The navigation bar includes links to the main page of Yang Lab (①) and the mTM-align server (②). Additionally, it offers examples (③④) demonstrating how our server operates, along with hits from various benchmark (⑤) tests. News (⑥) for mTM-algin are also provided.

① Go to the main page of Yang Lab.

② Go to the main page of mTM-align server.

③ Show an example of fast search of structure database.

④ Show an example of multiple protein structure alignment.

⑤ show the mTM-align result on benchmark homstrad, SABmark-sup, SABmark-twi, SISY-multiple, and D500.

⑥ show the news of mTM-align2.

Figure 2. The navigation bar of mTM-align server.

Input of Fast Search of Structure Database

This (①) is the input page for retrieving protein structure data. You can test our service using the provided example (②③). You have the option to input your protein sequence directly into the text box (④), upload a protein file (⑤), or load a protein (⑥) from the Protein Data Bank (PDB) or AlphaFold Database (AFDB). You can select the specific database (⑦) you wish to query and choose the retrieval mode (⑧) that best suits your needs. Additionally, you can provide your email address (⑨) to receive the retrieval results directly.

① Go to the input page of protein structure search.

② Input our example for monomer structure search.

③ Input our example for multimer structure search.

④ The text area for input.

⑤ Upload a protein structure file to retrieve.

⑥ Load a protein from PDB or AFDB.

⑦ Select the database that you want to retrieve.

⑧ Select the mode that you want to use.

⑨ If filled, the result will be sent to the email.

⑩ Give a name to this retrieval task.

⑪ Submit the protein and start the search task.

⑫ Reset the text area.

Figure 3. The submssion page of Fast Search of Structure Database.

Input of Multiple Protein Structure Alignment

This (①) is the input page for performing multiple protein structure alignment. You can download an example zip file(②) to validate our service. You can also upload (③) a compressed file containing multiple protein structures. Up to 20 structures are accepted.

① Go to the input page for multiple protein structure alignment.

② Download an example input for multiple protein structure alignment.

③ Upload a compressed file that containing multiple protein structures.

④ If filled, the result will be sent to the email.

⑤ Give a name to this task.

⑥ Submit the file and start the task.

⑦ Reset the uploaded file.

Figure 4. The submssion page of Multiple Protein Structure Alignment.

Output of Fast Search of Structure Database

This is the result page for retrieving protein structure data. The page displays proteins that share similar structures with your query. The organism (⑪) and molecule (⑩) information are also provided. You have the option to download (⑥) these proteins or align them with your query protein. Both pairwise (⑦) and multiple (④) structure alignments are supported. For multiple structure alignment, you should click on the checkbox (⑨) to choose the proteins that you want to align. Maybe you could click on ① to acquire the alignment for the top 10 result in this DB. Proteins from different DBs (③) are also accepted. Additionally, we offer prediction of the query's binding sites with the result structures. Please click on align (⑦) button to view the prediction.

① Performing multiple protein structure alignment for the top 10 proteins in the current DB tab.

② Download the result table.

③ Choose the DB that you want to view.

④ Performing multiple protein structure alignment for the proteins that you choose.

⑤ The ID of the hits, click on the ID will go to the detail page.

⑥ Download the result protein.

⑦ Align the query protein and the result protein.

⑧ The predicted similarity score of the result and the query.

⑨ Select the proteins that you want to perform multiple protein structure alignment.

⑩ The description for the molecule.

⑪ The organism that the protein comes from.

⑫ Download your query structure.

Figure 5. The output of Fast Search of Structure Database.

Output of Multiple Protein Structure Alignment

This is the result page for multiple protein structure alignment. We offer a visualization (①) for the aligned proteins, various viewing options (⑩⑪⑫) are provided. The alignment is also displayed at the residue level (②), and we provide a structure-based phylogenetic tree (③) for further analysis. Several indicators (④⑤⑥⑦⑧⑨) are shown to help you better understand the similarities between the proteins. Additionally, you can download (⑬) the files for the common core regions and the whole proteins.

① The visualization of the multiple protein structure alignment.

② Multiple protein structure alignment shown in residue level.

③ The phylogenetic tree computed base on structure similarity.

④ The length of common core regions.

⑤ The average pairwise RMSD in common core.

⑥ The average pairwise TM-score in common core.

⑦ The average pairwise length.

⑧ The average pairwise RMSD.

⑨ The average pairwise TM-score.

⑩ Show the common cores in magenta.

⑪ Show the proteins in different color.

⑫ Show the common core only in magenta.

⑬ Download files.

Figure 6. The output of Multiple Protein Structure Alignment.

Output of Pairwise Protein Structure Alignment

This is the result page for pairwise protein structure alignment. Firstly, we provide a visualization (①) of the aligned proteins, along with a residue-level structure alignment (⑫). Secondly, we predict the ligand (⑩) of the query protein based on the template. Clicking on the corresponding button to show the ligand along with the predicted binding residues (⑪). Additionally, various indicators (②③④⑤⑥⑦⑧⑨) are included to help understanding the similarity of the two proteins.

① The visualization of the pairwise protein structure alignment.

② The template that aligned with the query.

③ The length of the query.

④ The length of the template.

⑤ The number of residues that aligned.

⑥ The TM-score normalized by query.

⑦ The TM-score normalized by template.

⑧ The pairwise RMSD.

⑨ The sequence identity of the two protein.

⑩ The predicted ligand. Click on the ligand, it will be shown along with the protein.

⑪ The predicted binding residues. It is start with che chain id. Then the residue's one-letter notation along with its index are shown.

⑫ Structure alignment shown in residue level.

⑬ Download the structure.

Figure 7. The output of Pairwise Protein Structure Alignment.

References

R Dong, S Pan, Z Peng, Y Zhang, J Yang, mTM-align: a server for fast protein structure database search and multiple protein structure alignment, Nucleic Acids Research, 46: W380–W386 (2018).

R Dong, Z Peng, Y Zhang, J Yang, mTM-align: an algorithm for fast and accurate multiple protein structure alignment, Bioinformatics, 34: 1719-1725 (2018).