Zubia Mansoor, a second-year masters student in Simon Fraser University’s Department of Statistics and Actuarial Science, is the winner of DataJam 2020, a virtual one-day, hackathon-like event supported by CANSSI. This year’s topic focused on helping drive change towards sustainable oceans through technology.
DataJam 2020 consists of a day-long hack where hackers exercise what they have learned in teams, and then a networking event that includes a panel and keynote speakers, which included Nancy Heckman, Associate Scientific Director of CANSSI and Professor in the Statistics Department at the University of British Columbia. DataJam’s aim is to “foster a beginner-friendly environment and to increase the visibility of femme-identified individuals, WoC, and gender minorities in the fields of data science and technology”.
As the team lead at the Vancouver Datajam, Zubia and her team built an image classification algorithm that recognizes recyclables. The nine-member team were tasked to predict the category for different waste items using image classification techniques in machine learning and computer vision. For their project, they made use of secondary trash data maintained by Github users and supplemented it with pictures of trash that the team took themselves during the intense one-day hackathon. Their solution was to use pre-trained ResNet-19 features and fine tune them to train a classification network. This is very relevant in that classifiers like theirs can be used to correctly identify and sort recyclable items to reduce environmental contamination.
Zubia is also a part of the CANSSI Rapid Response Program project: Host genetics for SARS-CoV-2 severity and infection: Power and control, which is lead by Lloyd Elliott, Professor in SFU’s Department of Statistics and Actuarial Science.
This project aims to study the effects of gene mutation on severity. Any human genetic variation associated with SARS-CoV-2 infection after exposure, or with severity or outcome of COVID-19, may point to biological pathways that are involved in the disease course. Those genetic variations are thus of interest to vaccine and treatment development. Due to the way pandemics spread and phylogeography, genetic variation is confounded with ethnicity, socioeconomic indicators and occupation. In this project, one of the team’s main goals was to determine if COVID-19 and some neurodegenerative diseases share a common biological pathway as the literature suggests. Such a common pathway could also be explained away by the confounders listed above. To that end, Zubia and the team will examine genetic correlation between neurodegenerative diseases and COVID-19 having controlled for socioeconomic factors. From this work, the team will make recommendations for host genetics studies of SARS-CoV-2 and direct the genetics community towards important results.