Machine Learning Supercomputing Graphs and Statistics in Bioinformatics

Abstract

newline As we advance more into the world of quantum computing and machine learning, it is evident newlinethat we need standardization in using machine learning in the applied perspective. If we are newlineaiming for quantum machine learning to be applied to biomedical industry, we need to first make newlinesure that we have the framework ready for utilizing sophisticated supercomputing algorithms, newlinedata choice for next generation sequencing and its standardized usage, graph algorithms that can newlinework in a distributed computing environment to deal with large-scale data and give insights of newlineclusters and active entities or key entities involved, and then do downstream processing to get newlinebiological functions and disease relevance. This work largely addresses the pressing concern of newlineaddressing these issues and puts on table for practitioners a roadmap of how things should be newlinedone in terms of the value chain of particularly the next generation sequencing technologies newlinechoice, mapping and extraction of genomic variants, linking it up with expression of genes, newlinefinding clusters or hubs of activity, and then making statistical models or using machine learning newlinealgorithms such as random forest or logistic regression for instance to make statistical models newlinewhich can be used for scientific discoveries such as biomarkers for instance. Machine learning is newlinealso used for drug discovery or repositioning, and this method will also benefit using the next newlinegeneration sequencing technology and post-processing methods as discussed in this work