The project is from a Kaggle competition as the name shown.
COVID-19 pandemic broke out this year and has no sign of stop. People need effective vaccines to go back to normal life. Recently, we got great news from biotech companies Moderna and Pfizer. They have developed a new type of vaccine called the messenger Ribonucleic Acid (or mRNA) vaccine. Instead of injecting less lively viruses, mRNA strands directly enter body cells, send information of virus, and produce protein fight against them.
Compared to traditional ones, it is made from DNA templates rather than viruses. Also, it is easier to collect, because it is generated from an electronic sequence. Besides, the developing cycle is faster, typically within several weeks.
But the researchers found some issues that need to be solved. The key one is RNA molecules are spontaneously degraded and we are not sure which part of the backbone is most prone to being affected. And the currently available vaccines against COVID-19 must be shipped under intense refrigeration. However, shipping under high-level refrigeration limits the fraction of human beings these vaccines can reach.
The problem is from a Kaggle competition. The purpose is to leverage our data science expertise to develop models and design rules for RNA degradation. We are given a set of mRNA samples for training and finally, we are expected to predict degradation at each base position under different conditions(High pH level, high temperature, having magnesium element).
Qilun Lyu, Shang-Hao Huang
Kaggle: https://www.kaggle.com/c/stanford-covid-vaccine
YouTube Link: https://www.youtube.com/watch?v=OBkNZv-t9dY