Hello Bioinformatics lovers,

Tommy here. It was Mother's Day last weekend.

Did you call her?

I did, and I do it every week for > 1 hour.

I wrote a post about it here.

I am not going to teach you how to be a good son or daughter today:)

I am going to talk about Bioinformatics!

Let’s get one thing clear:

AI won’t replace you. But a biologist using AI? They absolutely will.

Especially in bioinformatics—where the questions never stop coming.

Back in 2013, when I got stuck on something technical (which was often), I’d head to Biostars or StackOverflow.

Sometimes I’d wait hours. Sometimes I’d wait days. Other times I never got an answer at all.

Now?

I ask ChatGPT. Or Claude. Or Gemini.

And I get a working answer in seconds.

That’s not magic. That’s leverage.

Even now, with 10+ years in the field, I still use AI every day.

Not because I’m lazy. But because it helps me think faster.

Someone recently asked me: Why do we calculate TSS enrichment scores in ATAC-seq?

Ten years ago, I had to dig through forums and PDFs to figure that out.

Today? I ask AI for a simple summary, then dive deeper.

Still curious? Here’s my full blog post from the pre-AI days: https://divingintogeneticsandgenomics.com/post/calculate-scatacseq-tss-enrichment-score/

I spent hours figuring out how to calculate the TSS score from ATACseq data.

Another one: Why do we use just 0.1% of input in a ChIP-seq experiment?

Simple answer: Input DNA = plentiful ChIP DNA = rare and specific

In ChIP, you pull down only the DNA bound to your target protein. It’s a tiny slice of the whole. AI explains that clearly now—no rabbit hole required.

These are the kinds of small but critical questions AI can now help you answer. Quickly. Clearly. Not always perfectly—but better than guessing.

Will AI replace biologists? No.

But if you’re a biologist not using AI? You’re already behind.

It’s never been easier to learn genomics, scripting, statistics, or experimental design. But only if you use the tools that exist.

Here’s what you can ask AI right now:

• “What’s the difference between PCA and UMAP?”
• “Write a bash loop to batch process FASTQ files”
• “Explain ChIP-seq vs ATAC-seq like I’m 10”
• “Create a Seurat filtering function for mitochondrial % > 10”

But here’s the rule:

Don’t get lazy. AI gives you speed. You bring the judgment. Always sanity-check its output.

Key takeaways:

• AI is a learning accelerator
• It won’t replace you, but it will amplify you
• Use it often, but use it wisely

Action items:

• Start asking AI real questions from your work
• Pair it with blog posts and scientific papers
• Share what you learn—it reinforces your growth

The future doesn’t belong to the smartest. It belongs to the curious. Especially the ones who ask good questions—and type well.

Posts from last week that you may find helpful

1. Be careful with outliers in your data, they can skew your correlation.
2. Your data is lying to you. Here’s how technical artifacts distort biology—and how to see the truth​
3. I had a 26GB TSV file. R choked. So I turned to UNIX. And it worked.
4. Bioinformatics gold is often found in the junk pile. Here’s how to mine it.
5. The Hidden Chaos Behind Beautiful Bioinformatics Figures.
6. BED files are an essential format for genomics data. You may be trapped if you are not careful 🧵
7. ​6 links on workflow to make your life easier 🧵
8. A year into bioinformatics, your code starts to work.But that’s also when it gets dangerous. Because now you can fool yourself.
9. Bioinformatics takes years to master. Not because it’s hard. But because so much of what matters… no one writes down.​

PS:

If you want to learn Bioinformatics, there are other ways that I can help:

1. My free YouTube Chatomics channel, make sure you subscribe to it.
2. I have many resources collected on my github here.
3. I have been writing blog posts for over 10 years https://divingintogeneticsandgenomics.com/​

Stay awesome!