Paper: PamaMultimer : A modular method for multimeric protein-protein interaction interface prediction,Yixiang Huang, Nan Zhao, Jiaqi Zhai, Xinqi Gong.
Abstract:
Although the protein monomer structure can be accurately predicted by alphafold2, the structure prediction of multimeric protein complex is still an urgent problem to be solved. Alphafold3 and alphafold multimer attempted to predict the structure of multimeric protein complexes and simulate the interaction between proteins and small molecules, but the results were not satisfactory. One of the main reasons for the above situation is that the interaction interface prediction is not accurate. To solve this problem, we proposed a multimeric protein complex interaction prediction method called PamaMultimer. In our research, we term the aggregation of amino acids in proximity to the central amino acid as a “patch”. By considering the solvent accessible surface area and internal contact area of the patch, we give the most likely interface patch and interface amino acids. It is worth mentioning that we used all the multimeric protein complexes (chain number greater than 2, monomers are proteins) that can be searched in the RCSB pdb library for testing, and the success rate and qualification rate of interface patch prediction reached 64.74% and 96.07%. The comprehensive and unomitted dataset strongly supports the accuracy and effectiveness of PamaMultimer. In addition, on the challenging antigen-antibody dataset, PamaMultimer predicted the top 10 binding sites with 48% accuracy. It has obvious advantages in comparison with the state-of-the-art methods. Different from the method of predicting amino acid binding sites, we use the patch as a whole to predict the interface. This helps to identify clusters of amino acids that work together to perform a function, which may help biologists make valuable discoveries. The code and data are available at https://github.com/hyx-1/PamaMultimer.
Introduction:
This is a online server to implement PamaMultimer, giving a series of multimeric protein complex residue patchs which may contain possible binding sites of these proteins.
Usage For Server: [NOTICE]: you must input your email address in the [email] textarea to receive the results !! Then you can click the [upload] button to submit a single pdb file or a zip file containing pdb files below, then click the [Submit] button. When your email address, job-ID and input file display in the text box below, you can click the [Run] botton and wait for a while. Otherwise, you may need to wait for a period of time or contact the Administrator of this website in Contact us. If you want to submit a new file, click the [Reset] button and repeat the steps above. The result file is a zip file. Note that the zip file includes different {pdb 's name}_result.txt files.
Get the Code:
If you would like to download the source code, you can visit this website:
How to read the result file:
We divide the results based on each chain of the protein and select the top 3 highest-scoring patches in each chain.These patches are considered to be the most likely to contain protein binding sites. In visualization file, we utilize specific color to label the top patches for users to see. In addition, The 'Repo_result' folder gives the number of surface patches per chain, the number of interface patches and the ranking of interface patches with the highest CSAIA value in surface patches.
Contact us:
If you have any questions during operation, you can contact us through 2021103691@ruc.edu.cn.
Try with Example:
We provide the example files below for you to try using the PamaMultimer server. The files include both a zip file which contains two dimers' pdb files and three dimers' pdb files.
You must provide your email address to receive the results !!! Then Click [Upload file] and then Click [Submit] !
