Molecular Dynamic Simulation of Neurexin1α Mutations Associated with Mental Disorder.

Neurexin1 gene is essential for formulating synaptic cell adhesion to establish synapses. In a previous work, 38 SNPs in Neurexin1 recoded in mental disorder patients have been collected. Five computational prediction tools have been used to predict the effect of SNPs on protein function and stability. Only four SNPs in Neurexin1α have deleterious prediction results from at least four tools. The current work aims to use molecular dynamic simulation (MD) to study the effects of the four mutations on Neurexin1α both on the whole protein as well as identifying affected domains by mutations. A protein model that consists of five domains out of six domains in the real protein was used; missing residues were added, and model was tested for quality. The MD experiment has last for 1.5 µs where four parameters have been used for studying the whole protein in addition to three more parameters for the domain analysis. The whole protein study has shown that two mutations E427I for Autism and R525C for non-syndromic intellectual disability (NSID) have distinctive behavior across the four used parameters. Domain study has confirmed the previous results where the five domains of R525C have acted differently from wild type (WT), while E427I has acted differently for four domains from wild type. The other two mutations D104H and G379E have three domains that only acted differently from wild type. The fourth domain of all mutations has an obvious distinctive behavior from wild type. Further study of E427I and R525C mutations can lead to better understanding of autism and NSID.

Improving classical scoring functions using random forest: The non-additivity of free energy terms' contributions in binding.

Despite recent efforts to improve the scoring performance of scoring functions, accurately predicting the binding affinity is still a challenging task. Therefore, different approaches were tried to improve the prediction performance of four scoring functions (x-score, vina, autodock, and rf-score) by substituting the linear regression model of classical scoring function by random forest to examine the performance improvement if an additive functional form is not imposed, and by combining different scoring functions into hybrid ones. The datasets were derived from the PDBbind-CN database version 2016. When evaluating the original scoring functions on the generic dataset, rf-score has outperformed classical scoring functions, which shows the superiority of descriptor-based scoring functions. Substituting linear regression as a linear model by random forest as a nonlinear model had largely improved the scoring performance of autodock and vina while x-score had only a slight performance increase. All hybrid scoring functions had only a slight improvement-if any-on both of the combined scoring functions, which is not worth the slower calculation time.