Machine Learning in Multidimensional Biomarker Design: A Milestone in Precision Medicine - A Systematic Review

Raajasiri Iyengar, Gaurav ,
2020 Acta Scientific Dental Scienecs  
Introduction Beginning with the completion of a milestone in our genome sequencing endeavor, the Human Genome Project (HGP), we have now evolved to complex sequencing technologies known as highthroughput sequencing or next-generation sequencing (NGS). As a result of this dynamic evolution, we now have enormous genomic datasets have been mined to identify both drug targets and bio-Background: With the emerging era of precision medicine and high-throughput sequencing technologies, huge amount of
more » ... omics' data has been gathered. Data interpretation remains a challenge due to the large and evolving magnitude of the dataset [3]. Precision medicine not only includes targeted therapeutics, but also necessitates 'precision diagnostics'. Rational design of multidimensional biomarkers is the keystone of precision diagnostics, wherein multiple biomarkers are cumulatively assessed using computational techniques to yield specific patterns [1] . Patterns thus obtained are analyzed with Machine learning algorithms to arrive at a diagnosis that is both reliable and accurate. Aim of the Study: To determine the role of Machine Learning in the rational design of multidimensional biomarkers. Research Question: Can Machine Learning enable rational design of multidimensional biomarkers which serve as molecular signatures for specific disease states? Materials and Methods: With the Medline database and Cochrane Collaboration taken as a source for authenticated scientific research data, 45 articles were selected having undergone randomized control trial. Out of these, 14 articles (studies) were chosen which met the criteria for systematic review. Results and Conclusion: Machine learning enables identification of molecular signatures of specific disease states, by cumulative interpretation of multiple biomarkers simultaneously. ML Algorithms can discover higher-order interactions among biomarkers, greatly improving the diagnostic performance.
doi:10.31080/asds.2020.04.0957 fatcat:sorebztv6fblrk2xyzecnlwvki