Disclaimer: This project idea is the intellectual property of the ByteSize Team.
Team: ByteSize (Participating team of 2018)
Members:
Asif Shikder and Lina Deng, New York City College of Technology
Naman Pujari, City College of New York
Sumaiya Urboshe, Queens College
Link to the Business Case Analysis.
Overview of the Project – The Department of Health and Mental Hygiene (DOHMH) and Center of Disease Control (CDC) have been pushing for the implementation of Whole Genome Sequencing (WGS) across all Public Health Laboratories (PHL) in the United States. This initiative will allow for better analysis of foodborne outbreaks and faster development of solutions to combat them. However, the current method of conducting an analysis is extremely time- consuming, expensive, and subjective. As a result, CDC is unable to fully utilize the potential of WGS technology.
This project proposes a Watson Machine Learning (WML) technology, to speed up this process by 7 times, and cut down cost by 50%.
Problem – The current process for WGS at the PHL is time-consuming, expensive and subjective. Scientists need to manually review and compare numerous modules, each covering different sequence-indicative numeric, to make judgments on their validities.
Approach to address the problem – Implementation of the Watson Machine Learning for analyzing sequences generated by WGS will help quantify each module by setting a quality-indicative score system.
Benefits and Impacts – The solution will significantly reduce the turnaround time per sequence from 28 days down to just a little over 4 days. It will also cut down the labor cost of each sequence from $1,248.00, to $617.74 for each sequencing runs.
We are able to cut down the cost of per genome sequencing runs from $1, 2480.00 to $617.74, in addition to providing a 7x increase in turn-around time and total savings of $1,000,000 in just 5 months.