Looking at bioinformaticians’ profiles these days, you'd think everyone has decades of experience in cutting-edge single-cell and AI-driven bioinformatics. But something’s missing… 👇
Comments
Log in with your Bluesky account to leave a comment
1/ It feels like basic bioinformatics skills aren’t getting enough love anymore. Everyone’s talking about their deep learning pipelines and generative AI models, but what about fundamental skills like exploratory data analysis (EDA) and data sanity checks? 🤷♂️
2/ Working with bio data isn’t always about fancy models. It’s often about handling messy real-world data—doing quality control, normalization, and making sure the data makes sense before even thinking about machine learning.
3/ The field didn’t start with deep learning models on single-cell data or transformer models for genomics. It started with solid EDA, data wrangling, and a deep understanding of the data to spot biological patterns.
4/ Unfortunately, it’s becoming harder to find bioinformaticians who can double-check results, question their data, and say, “Wait, does this make sense biologically?” instead of rushing to fit the latest deep model.
5/ Sure, it’s great to have machine learning skills (I am going over the course on https://fast.ai too), but bioinformatics isn’t just about flashy models. It's about knowing when to trust your data and when to question it.
6/ A lot of the breakthroughs in bioinformatics came from people who asked simple, smart questions about their data. You don’t need a PhD in AI to make an impact—you need common sense, curiosity, and a solid foundation in basic data skills.
Comments