Advanced Survival Analysis Reading

While not exactly a project yet, my first task is to self-study advanced survival analysis and counting process theory (mostly from Counting Processes and Survival Analysis by Thomas R. Fleming and David P. Harrington). My first main project is to investigate the power and sample size for a combination study of individual-randomized and cluster-randomized clinical trials. This method mostly surrounds investigating the efficacy of vaccines.

I am including this incomplete section so that I can post my notes on advanced survival as someone who hates books that handwave things. These notes are built upon notes from Duke and UPenn's advanced survival analysis course and notes collected from Dr. Tsiatis.

View only notes: https://www.overleaf.com/read/ffwrqbrzbqjj#0e6d88

Undergraduate Research @

University of Toronto

Basket Trials Designs @ The Centre for Addiction and Mental Health

At the Centre for Addiction and Mental Health (CAMH) I worked in the biostatistics core service, where I attended biostatistics meetings and discussed new methodologies being developed to help solve issues prompted by clinicians. My work began after discussing the use of psilocybin against obsessive-compulsive disorder, body dysmorphic disorder, and anorexia nervosa in a basket trial. I am actively working on completing the associated paper and deriving the necessary formulae. We hope to publish to Clinical Trials.

JSM_Presentation CM230709 MAIN.pptx

In-Silico Protein Engineering @ Pardee Lab (Dept. of Pharmacy)

At the Pardee Lab in the Department of Pharmacy, I worked on the development of a novel machine-learning algorithm/pipeline to discover novel amino acid sequences for a targeted purpose. I spent a year developing the machine learning pipeline, which yielded a 60% accuracy, approximately 20% higher than existing methods on 25 training points on testing data. In my final couple of months, I led a team of 3 more undergraduates to wrap up model development. I am no longer employed at the lab, but current efforts are focused on building a high-throughput system that can screen proteins and improve the model's accuracy.

Final Report (no mention of wet lab results): https://drive.google.com/file/d/1Y6j19-G55BfxreJUkHyOVI_ZNSE2TZ22/view?usp=sharing

Page updated

Report abuse