Study of Evolution By Automating Hardy Weinberg Equilibrium With Machine Learning Techniques In TensorFlow And Keras

Evolution transpires in organism populations and includes population diversity, heredity and varying survival. One way to study evolution is to study how the frequency of a population's allele varies from generation to generation. An allele is a type of gene variant. In other words, rather than...

Full description

Saved in:
Bibliographic Details
Published in2020 Advanced Computing and Communication Technologies for High Performance Applications (ACCTHPA) pp. 14 - 19
Main Authors Balan, Karthika, Santora, Michael, Faied, Mariam, Carmona-Galindo, Victor D.
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.07.2020
Subjects
ADAM Optimization
Allele Frequency
Biological system modeling
Evolution
Genetics
Hardy-Weinberg principle
Keras
Machine Learning
Mathematical model
Optimization
Python
Sociology
Statistics
Tensorflow
Training
Online AccessGet full text

Cover

More Information
Summary:Evolution transpires in organism populations and includes population diversity, heredity and varying survival. One way to study evolution is to study how the frequency of a population's allele varies from generation to generation. An allele is a type of gene variant. In other words, rather than looking at two parental species, evolution should be investigated by looking at the population as a whole. The only way to do this is therefore to investigate how allele frequencies shift in populations and how these shifts will predict what will happen to a population in the future. Throughout biology, there are basic equations which are used to model this theory basically termed as Hardy-Weinberg principle. This paper describes a new method that integrates Ecology and Engineering by automating the Hardy-Weinberg model using ADAM Optimization technique. The neural network trained based on Keras and Tensorflow has been deployed using python for the study of evolution. The developed model is tested with standard allele frequency database "ALFRED" to prove the efficiency.
DOI:10.1109/ACCTHPA49271.2020.9213202