“The C3G internship allowed me to expand my (bio) informatic skill to a new professional level. Personally this helped me to understand how I want to focus my skills and job further on. Moreover, I learned a whole new set of skills that will allow me to expand my job opportunities.”

"During my summer interning at C3G in Montreal, I worked on developing a logging system aimed at helping users debug and monitor instances of Bento. The core technologies used were Grafana and Loki. Grafana was instrumental for visualizing and analyzing logs, while Loki helped with aggregating and querying log data efficiently. This exposure to modern logging tools deepened my understanding of how to manage complex systems effectively. In addition, integrating these tools into the existing framework gave me broad experience with relevant technologies such as Docker, React, Python, and NGINX, all in a production environment. Overall, it was an awesome learning experience and I wish the best for C3G in the future!"

“For my internship, I was tasked with creating an interactive dashboard that showed the evolution of different run sequencing metrics over time. This opportunity enabled me to enhance my proficiency in Python and SQL, while also acquiring new skills in R Shiny, Docker, and Bash within an HPC environment. Given my background in biology and a year of prior bioinformatics training, this experience allowed me to apply the knowledge I had gained in the classroom and explore various aspects of application development that were previously unfamiliar to me.”

“I am a third-year McGill student studying mathematics, with an interest in linguistics and law. During my internship at C3G, I was tasked with developing SD4H’s privacy and security framework and assisting in writing SD4H’s policy on data protection. To do this, I collaborated with professionals from technical, legal and administrative backgrounds. I am looking forward to continuing to work at C3G in the months to come.”

“During my extensive engagement at C3G, spanning over summers, falls, and winters, I played a pivotal role in the Bento project team as a full-stack developer. My primary responsibility revolved around frontend development. I was instrumental in upgrading and adapting the frontend interfaces to meet specific requirements, yet I always maintained a balance by ensuring my creative autonomy wasn’t compromised. The culmination of these efforts was the redesign of “Bento-public.” This overhaul wasn’t just a cosmetic change; it involved improving user experience, optimizing performance, and ensuring responsiveness across devices. Post-revamp, it’s now hailed as a benchmark, often referred to as the template for designing other frontends within our ecosystem. As I delve into my current endeavours, I am immersing myself in designing an innovative frontend appearance. Utilizing Figma as my design tool, I’m crafting detailed prototypes, ensuring they align seamlessly with our technical and aesthetic specifications. One of the standout features of my role was how closely the frontend was interwoven with the backend. This interdependency provided a unique challenge and learning curve. Thanks to the unparalleled freedom I was granted, I could delve deep into multiple layers of the tech stack. This experience enriched my understanding of Django’s framework intricacies, powered by Python. I honed my skills in structuring and querying databases using SQL, learned the dynamics of the Go programming language, and familiarized myself with containerization principles using Docker. These hands-on experiences didn’t just augment my practical skills, but they also provided me with a holistic view of full-stack development, from ideation to deployment.”

“For my internship, I have been working primarily on the Freezeman project, a sample tracking application. The project uses React (with Javascript) to build web pages and Django (with Python) for the backend. My primary role in the internship was to implement new features and fix existing features in the website. The first contribution I have made was to produce an interactive graph that lets users explore the processes that a sample goes through such as extraction. The most significant project I worked on was reducing the time to export information of samples in a CSV file which made it feasible for users to export 200 000 samples and more in less than a minute. With the help of the Magic team, and through project exploration, I began to understand how their system works. By understanding Freezeman, I also started to understand basic concepts related to genomic sequencing processes such as normalization, pooling, and many others.”

“For me, being an intern at C3G has been a unique chance to discover new ways of working and also to learn a lot about Genpipes and bioinformatics in general. Having a bilingual team was also a good way to improve my English. This internship has also been a great opportunity to discover Montreal and Canada in general!”

“My internship at C3G came at the end of my master’s degree in bioinformatics in France. I had two main missions during these 6 months. The first was to build a custom snakemake pipeline to use NGSCheckMate to check that samples from the same patient correlate well together, or to identify swaps or contamination at an early stage. The tool uses allelic frequencies to compute correlations between two samples, taking into account the 20,000 most variable positions in the human genome. My second mission was to create an R Shiny interface to enable users to explore the results of Single Cell analyzes themselves. These are generally analyzed with a lot of expression values for each gene, and there can be a lot of them. Having an interface for testing multiple graphs and genes is useful and helps users better understand their results. In particular, it gave me a better understanding of network architecture. I was really grateful to be able to have this experience at C3G to have a better technical experience and to be backstage in a bioinformatics platform. I was able to meet a lot of very nice people with very wide-ranging and complementary skills in computer science and biology.”

“During my summer with the C3G’s TechDev team, I was tasked with benchmarking various Methyl-seq aligners to test for speed, memory use and compatibility with the expected output with the GenPipes pipeline. GenPipes has been using Bismark to align WGBS short reads, which is infamous for being slow, albeit accurate. During my time, I successfully benchmarked a handful of potential Methyl-seq aligners, eventually finding GemBS. This software had a large improvement of speed as well as a marginal improvement in overall accuracy. At the end of my internship, I was able to begin implementing GemBS into GenPipes pipeline to replace Bismark, as well as add various other Methylseq tools to the pipeline such as MethylDackel. My summer internship with C3G was an amazing experience. I was able to learn technical skills such as working with Unix, Python development, working with a HPC environment and working with various Bioinformatic file formats and tools. I also had a chance to hone my theoretical knowledge of bioinformatics and genomics with the application of what I’ve learned during my degree, as well as weekly Journal Clubs!”

“During my internship at C3G, I had the opportunity to work on multiple projects doing hands-on bioinformatics. The first project was to look at a transgenic mouse’s genome sequencing data for an insertion/deletion event which wasn’t picked up on by the usual detection methods. Afterwards, I compared the differences between using the GRCh37 and GRCh38 genome references by applying the GenPipes RNA-seq pipeline on previously analyzed data. I made a few interesting discoveries that proved useful to C3G in order to improve their analyses. Next, I worked on a project with the goal of replicating results from a previous study looking into optimizing the sequencing and detection of cassava mosaic viruses (CMV) present in African crops (TreeLab).The end goal of said project being to be able to detect CMV with a laptop unconnected from the Internet directly in the fields.”

“Throughout my summer internship at the C3G, I worked on two main projects. First, I developed web based data visualization tools for a federated genomics database. These tools helped users visualize genomic variants and the information embedded in them. The second project I worked on was to implement a decision tree to guide researchers in protecting the privacy and confidentiality of the human health-related datasets they process. Some of the technologies I used during my internship were: Javascript, Python and React. Since I did not have any prior experience with genomics I learned a lot about bioinformatics and the way the human genome works (specially genomic variants). In addition, I learned about the different regulation laws that protect personal genomic information.”

“I was tasked with organizing massive sets of epigenomic data in accordance with IHEC standards during my internship. When I arrived, the data was very loosely organized by consortium, all of which had different naming conventions and internal organizational structures. There were three steps to organize the data. First, I wrote a script to scrape information for the IHEC database and store it as a master metadata file. With IHEC IDs and consortium IDs linked, I then wrote a program to move data files into a structure that better followed the international standard. The most interesting part of my internship was writing a search program that allowed users to quickly locate files based on dataset characteristics, such as donor age, tissue type, or experiment type. I quickly learned that even the best file organizational system would not work for everyone, so the search function was key to meet the goals of my internship. I loved working with C3G. I had an opportunity for the first time to use powerful computing clusters, became more familiar with working in a UNIX environment, and thoroughly developed my Python skills.”

“Over the course of my internship with C3G, I was involved with a few key projects. More specifically, I was tasked with generating synthetic FHIR datasets using FHIR APIs. These datasets were created with the goal of ingesting them into the metadata service and testing the ingestion algorithm. All synthetic datasets that I generated were published on C3G’s public repository, so other developers and researchers will have access to already prepared datasets. A large aspect of my internship involved mapping across data standards. I worked on mapping the GA4GH Phenopackets standard used by the metadata service to the mCODE standard. This mapping will be used in future C3G projects that will benefit from incorporating specific mCODE data elements. Finally, I was tasked with mapping data elements collected by a COVID-19 initiative, CanCOGen, to the Phenopackets schema. Ultimately, the mapping I completed was sufficient for building a synthetic dataset for a prototype to be used for the CanCOGen project. My time as an intern with C3G has been very hands-on and educational. I learned a lot through my work with APIs and extensive documentation of data standards, as well as through interactions with my colleagues. As I have a more clinical background, rather than a computer science background, my internship has made me eager to continue to strengthen my programming skills.”

“During my internship, I worked on the development of a new pipeline for GenPipes called EpiQC. EpiQC can be used on C3G’s computer clusters. The pipeline’s objective is to assess the quality of a ChIP-Seq signal track (bigwig) dataset and determine if it should be used or not. I implemented different metrics to verify the quality of those bigwig files using 3rd party tools such as BigWigInfo, ChromImpute and EpiGeEC and also coded multiple python scripts to facilitate the use of the pipeline. After having tested the pipeline on small datasets (~100 files), I ran it on the whole ChIP-Seq data available through the IHEC data portal (~2500 files) to see how the pipeline runs on a big dataset. While most of the features worked, I had some issues with ChromImpute. I successfully trained ChromImpute on the whole dataset but could not impute the predicted files. I learned a lot about computer clusters since I had no previous experience with them and how to handle large datasets, and while coding the pipeline and various scripts, I further improved my skill in using python.”

“Summer of 2019, I worked on metagenomics analysis. In particular, I installed the whole genome shotgun (WGS) pipeline, MOCAT2, on C3G’s computing clusters. Since it did not work out of the box, we did a lot of debugging and wrote our own wrapper scripts that were tested to work on C3G’s servers (since the ones shipped with MOCAT2 were depreciated). Additionally, we modified some of the MOCAT2 code to bring the run time of one of the pipeline’s steps down from several hours (and sometimes days, depending on the sample size) to only minutes and reduced RAM usage from 250 – 500GB of RAM down to less than 1GB of RAM. We used MOCAT2 to analyze 156 stool samples to examine the relationship between the gut microbiome and fibromyalgia. Since I had no bioinformatics experience and minimal computer science experience, I learned a lot during this internship, including genome sequencing and the importance of the gut microbiome, how to debug large pipelines and write my own wrapper scripts, how to process very large datasets, and how to use computing clusters.”

“During my time as a web development intern at C3G, I worked to improve the GenPipes Dashboard for viewing pipeline execution. I got a lot of great experience working with modern frameworks in the frontend and backend, such as React, Redux, Node.js, and Express. My main focus throughout the internship was to improve the performance of the website, as well as add new features to provide a better user experience. Performance enhancements include adding an efficient sample search feature, as well as lazy-loading and virtual scrolling for samples. I also added a graphical view of sample execution, created a page to view project statistics, and provided a general cleanup of the user interface. Throughout the internship, I got a lot of experience with modern web frameworks, as well as learned the best practices for keeping code clean and concise.”

“As a software development intern at C3G, I worked on three projects: a tool to look at the research output of Genome Canada, a website which displays research connections between professors at McGill and especially the McGill initiative in Computational Medicine, and a browser for data produced by the MHcut tool (a collaboration between Dr. Bourque’s lab and the Woltgen lab at Kyoto University.) During my internship, I primarily used Python to write scripts and web service backends, and D3 to produce interactive data-driven JavaScript-based front-ends.”