Everyone Struggles

What were your challenges during the contribution stage?

During the contribution stage, there was a model I was to test. The model was about “Human Liver Microsomal Stability.” I needed to understand the model so I could interpret the output of the AI model. The model was obtained from the research paper. To understand the model, I did some research on Google and read the research paper. The model predicts the “Human Liver Microsomal Stability” and the “Rat Liver Microsomal Stability.”

The “Human Liver Microsomal Stability” model assesses how stable a chemical compound is in the human body, specifically in the liver. The stability of a compound is crucial in drug development because it affects how long the compound remains in the body before it is broken down or metabolized. The model helps predict whether a chemical compound will be stable or unstable in the human liver and rat liver, providing valuable information for drug development and understanding how drugs are metabolized in the body.

What other challenges did you face in terms of technical tools?

I wanted to serve the model from the Ersilia Model Hub using Docker. I’ve never used Docker before. It’s been on my to-learn list. I saw some YouTube videos on how to use the Docker Hub to download Docker images. I was able to serve a docker image containing the machine learning model successfully. I am also learning more about Docker on DataCamp.

What tools do you use?

I read the documentation online, for example, the Scikit Learn documentation. I make use of YouTube to see video tutorials. I also use ChatGPT by asking it to explain technical concepts in a simple way I can understand. I read research papers to understand the context of what I’m working on. I also read blog posts. As a last resort, I ask my friends and mentors for help. I like doing my research first. If the problem persists after doing my research and using all these tools, I reach out for help. 

How do Your Mentors Help?

My mentor has been really helpful. When she explains things, she also takes her time to explain some of the concepts in biomedicine and drug discovery. It helps me understand the models I’m working on. I also ask a lot of questions. I am new to the industry and I find it fascinating. I’m curious and I want to learn more.

What would you tell someone worried about asking questions?

I used to be worried and scared about asking too many questions. It made me feel like I wasn’t competent or I did not know what I was doing. But it’s the opposite. As a data scientist, I’ve learned the importance of asking questions. The course “Ask Questions to Make Data-Driven Decisions” by Google on Coursera taught me the importance of asking questions.

It’s okay to ask a lot of questions. It shows you want to understand what you are doing, so you don’t end up doing the wrong thing. As a freelancer, I ask my clients a lot of questions to know what they want and to make sure I deliver the right result.