Director of bioinformatics at AstraZeneca. subscribe to my youtube channel @chatomics. On my way to helping 1 million people learn bioinformatics. Educator, Biotech, single cell. Also talks about leadership.
tommytang.bio.link
Ming Tommy Tang
Loading...
The job is not disappearing. It is shifting.
Less time clicking through Cell Ranger and STAR. More time designing the experiment, picking the right controls, and catching the moment the model confidently produced nonsense.
That second list is harder and where the value is.
The bioinformaticians who do well over the next five years are not the ones who refuse to touch AI.
They are the ones who use it every day, on hard problems, and still know exactly when it is wrong.
Paper: www.nature.com/articles/s4...
I hope you've found this post helpful.
Follow me for more.
Subscribe to my FREE newsletter chatomics to learn bioinformatics divingintogeneticsandgenomics.ck.page/profile
Ming Tommy Tang
Ming Tommy Tang
AI is an accelerant. It is not a replacement.
The paper's argument: AI value collapses without expert judgment in four places. Study design. Data curation. Interpretation. Governance.
Skip any of them and you ship fast garbage faster.
Ming Tommy Tang
I see this every week.
Someone runs a single-cell pipeline through an AI agent. Clean UMAP, cell types labeled, figure ready for the slide deck.
Then I ask why their tumor samples cluster by sequencing batch instead of by subtypes.
Silence.
Someone asks me every other week if AI is going to replace bioinformaticians.
A new npj Digital Medicine paper finally said what I have been telling them for two years.
What AI cannot do for you:
Tell you that your batch correction wiped out the biology, not the noise.
Pause before enriching a "novel" pathway that anyone who has looked at TCGA twice would call an artifact.
You have to know.