Description
Alternative splicing allows for gene products having various functions and increases protein diversity. The percent spliced in (PSI) indicates the efficiency of splicing a specific exon into the transcript population of a gene(1). In this project we attempt to predict the PSI of a set of exons using various gene regulators using a deep neural network and a shallow wide neural network. A better performance from neural networks is achieved thorough hyper parameter search which is capped by the computational power.