How we hacked two Med.Hack(+) challenges
The desire to help, learn, and grow unite the issues of remote communities’ access to healthcare and the fight against SaRS-CoV2
Vendasta is involved in several community programs that help children learn how to code, deliver meals during yearly holidays, make charitable contributions, and help promote a strong technology ecosystem in Canada. To bolster its involvement in the community, Vendasta provided its R&D division the opportunity to take company time to participate in Med.Hack(+) 2020! R&D already hosts its own hackathon every 8 weeks, so this was an additional growth opportunity for developers, and an extra chance for Vendasta to be involved in the community!
Med.Hack(+) is a hackathon that aims to provide individuals with the opportunity to collaborate with healthcare professionals in order to tackle problems in healthcare through applications of technology. At a high level, Med.Hack(+) “brings together problems in healthcare and problem solvers who can build solutions through the use of technology.”
While typically theme-less, this year’s Med.Hack(+) focused on the grand issue presented by the current pandemic, while still accepting challenges for other problems. Vendasta had several developers join the hackathon to work on two interesting challenges: (1) remote communities’ access to healthcare treatments, and (2) SaRS-CoV2 drug repurposing and bootstrapping.
Remote communities access to healthcare
The problem
Access of remote communities to healthcare is a general problem that affects the province of Saskatchewan, Canada. The organization of Métis Nation — Saskatchewan presented this challenge at Med.Hack(+) by showcasing a case study of individuals who are affected by cancer and live in northern communities. Individuals undergoing cancer treatment require access to the Saskatchewan Cancer Center, which is located in Saskatoon, Saskatchewan, Canada (the home of Vendasta).
The solution
A group of healthcare practitioners, representatives of Métis Nation, and a Vendasta developer proposed the development of a progressive web application (PWA) that would help pair drivers and patients from remote communities. In the context of this PWA, drivers are the people who are leaving remote communities to drive to a city such as Saskatoon, Saskatchewan, and patients are the ones who need access to the city.
The solution involved developing a FireBase application that allowed signups from both sides — the drivers, and the patients — in order to pair them up. The reason behind using a FireBase application was to facilitate access on mobile and desktop devices to maximize the likelihood of remote patients having the opportunity to use the service. The app allowed sign ups with Google accounts, a way to indicate destinations (from both the driver and the patient), a way for the drivers to accept patients, a way for patients to request a ride, and, finally, a mechanism for pairing patients and drivers. In addition to pairing drivers and patients, the application also plans to integrate with the eHealth directory of Saskatchewan to potentially facilitate ride payments through health coverage, record-keeping of the pairs on patient health records, and verification/audit for patient safety and driver verification.
This project ended up being designated as the winner of this year’s Med.Hack(+) edition. While the reasons were not published, we can hypothesize that the PWA approach presents an immediately actionable approach towards facilitating remote communities access to healthcare.
SaRS-CoV2 drug repurposing and bootstrapping
The problem
According to the World Health Organization, over 1M people have died of COVID-19 worldwide. Similarly, Canada is approaching 10K due to a second wave of cases. Approaches such as social distancing and nation-wide lockdowns led to severe economic impacts around the world and, while these measures have slowed down the spread of the virus, those who contract the virus still do not have any reliable treatment available. There is a small number of drugs (Dexamethasone and Remdesivir) available upon physician request, but they are either in limited supply, or can only be used for severe cases. Therefore, expanding the panel of drugs available would increase the number of options for patients. New, or repurposed drugs, can decrease the time to market of available treatments for COVID-19. Increased availability of antiviral drugs can help prevent the spread of SaRS-CoV2, which enhances the safety of communities, ultimately saving lives. In addition, increased availability can help nations alleviate some of their lockdown and social distancing measures, which will have positive economic impacts.
The solution
A group consisting of several developers from Vendasta, developers from other companies, and a vaccinology graduate student worked on identifying potential genetic drugs that could help bootstrap new drugs for the treatment of COVID-19. The first solution involved investigating the function of each SaRS-CoV2 gene to map the function to specific gene-drug interaction categories to sample existing drugs as candidates.
The solution involved downloading the SaRS-CoV2 genome using BioPython. The genome was used to extract the sequences of the SaRS-Cov2 genes. The functions of each gene was documented using existing NCBI SaRS-CoV2 resources and mapped to a corresponding gene-drug interaction in DGIDB. All the information was extracted using the DGIDB API and the BioPython Entrez package. In addition, the alignments were performed using the BioPython pairwise2 package. The top 5 alignments were chosen for each pair of SaRS-CoV2 gene and the gene from the documented gene-drug interaction from DGIDB, after sorting alignments based on the similarity score.
Since the group consisted of a considerable number of members, a second solution was pursued by a subset of the members. The second solution investigated the transcription factors that might be used by SaRS-CoV2 by performing a search for similar sequences in other viruses’ genetic makeup and a scan for drugs that can target those transcription factors. The NCBI Nucleotide BLAST was used to search for similar sequences to the SaRS-CoV2 genes. The family of Coronaviridae viruses was used to search for sequences, but the SaRS-CoV2 virus was excluded from the search. The identified sequences were saved in JSON format, extracted, and FIMO was used to search for potential transcription factors that might regulate SaRS-CoV2 genes. The JASPAR vertebrate database was used for searching transcription factors with FIMO. Similar to the first approach, the identified transcription factors were mapped to the known transcription factor drug interactions from DGIDB, which yielded additional candidate drugs.
Overall, the two approaches resulted in 422 candidate drugs that were briefly compared and verified against existing publications that document potential drug candidates based on approaches such as protein-protein interactions, and gene expression profiles.
Conclusion
Med.Hack(+) offered Vendasta the opportunity to be involved in very interesting and challenging problems that are affecting the province of Saskatchewan, and the world. Personally, I had the opportunity to work and share a weekend with amazing people — Vendastians, and non-Vendastians.
Vendasta helps small and medium-sized businesses around the world manage their online presence. A website is pivotal for managing online presence and performing eCommerce. My team is working on a website-hosting platform that faces numerous interesting scalability (millions of requests per day), performance (loading speed), and storage (read, and write speed) problems. Consider reaching out to me, or applying to work at Vendasta, if you’d like a challenge writing Go and working with Kubernetes, NGINX, Redis, and CloudSQL!
