Lian, D. | bioRxiv Science

bioRxiv2025

The Polymorphisms, Solvent Accessibility and Conservatism of Hepatitis C Virus Nonstructural 5B Protein

ABSTRACTThe polymorphisms of protein or protein family, that is, the divergences of amino acid and nucleotide sequences have provided much useful information on the divergent evolution of proteins. In this paper, we analyzed the polymorphisms of enzyme NS5B of HCV for which sequence variation among most isolates have been characterized and protein structures of the catalytic domain form of this enzyme are also known. For this protein, we found that solvent accessibility of residues in the protein structure is a strong predictor of whether or not an amino acid will be polymorphic and the residue variability. Apart from polymorphism, we found conservatism at every level among site is universal for this protein. We also found that purifying selection at different levels was strong in the forming of the polymorphisms and conservatism of this protein.

bioinformatics↗

bioRxiv2023

Prediction of Carbon Emissions in Guizhou Province-Based on Different Neural Network Models

Global warming caused by greenhouse gas emissions has become a major challenge facing people all over the world. The study of regional human activities and their impacts on carbon emissions is of great significance to achieve the ambitious goal of carbon neutrality and sustainable economic development. Guizhou Province is a typical karst area in China, and its energy consumption is mainly based on fossil fuels.Therefore, it is necessary to predict and analyze its carbon emissions. In this paper, BP neural network and extreme learning machine (ELM) model, which have the advantage of nonlinear processing, will be used to predict the carbon emissions of Guizhou Province from 2020 to 2040. Based on the energy consumption data of Guizhou Province, the carbon emissions of Guizhou Province are calculated by using the conversion method and the inventory compilation method. The data show that the carbon emissions of Guizhou Province show an "S" growth trend; In this paper, 12 influencing factors of carbon emissions are selected, and five influencing factors with larger correlation are screened out by using grey correlation analysis method, and the prediction model of carbon emissions in Guizhou Province is established and simulated, and the prediction performance of BP neural network, ELM and WOA-ELM are compared respectively. Compared with ELM model and BP neural network model, the prediction accuracy of WOA-ELM model is higher; Finally, three development scenarios of carbon emissions are set by scenario analysis, which are baseline scenario, high-speed scenario and low-carbon scenario. On this basis, the size and time of peak carbon emissions in Liaoning Province from 2020 to 2040 are predicted based on WOA-ELM model. The results show that the peak value of carbon dioxide in the low carbon scenario is up to 0.98 million tons 31294 in 2033, the peak value of carbon emissions in the high speed scenario is up to 0.37 million tons 30251 in 2036, and the peak value of carbon emissions in the baseline scenario is up to 0.61 million tons 26243 in 2038. Based on the peak time and prediction results of carbon emissions under the three scenarios, the main factors contributing to the reduction of carbon emissions in Guizhou Province are analyzed, and important data basis is provided for energy conservation and emission reduction in Guizhou Province.

ecology↗