Hello Bioinformatics lovers,

Again, a week has passed. I never felt that time had passed so fast until I saw how big my three kids had grown.

I can barely carry my 8-year-old daughter (almost 60 lbs!).

The world has also changed rapidly since the AI buzz resurged after chatGPT was first released. How does this relate to our bioinformatics learning?

We should embrace those AI tools to help us learn faster!

AI will not replace a bioinformatician, but someone else who uses AI will

Now, I am not talking about applying AI for bioinformatics and drug discovery. I have a blog post on it.

I am talking about how to use AI tools to speed up our learning.

I use chatGPT extensively to prototype code and learn a topic. For example,

I asked it to give me a low-level understanding of Seurat PCA/CCA label transfer for scRNAseq by Mutual Nearest Neighbor, and I wrote two posts:

​How PCA projection and cell label transfer work in Seurat​

​How CCA alignment and cell label transfer work in Seurat​

I won't say it is easy. I spent at least 6-8 hours on each post. And I went to sleep at 2 am for several nights.

There will be errors with the ChatGPT code. and it is NORMAL.

Especially in R, the matrix is represented by gene x cell, but most math was done by cell x gene matrix so you will need to transpose the matrix to feed into many of the functions in R.

I enjoyed it even though I bang my head on the table when I see the results do not make any sense! (Protip, use a dataset that you understand so you can evaluate the output of the code!)

Learning through suffering is a new term I recently learned.

It is through suffering that you learn the most and have the deepest understanding of a problem :)

I hope you start using those tools to 10x your learning of bioinformatics.

• ChatGPT
• Claude
• Perplexity
• github copilot and my others

Happy Learning!

Tommy aka, Crazyhottommy

PS:

In my recent video, I showed you how to liftover bed files and bigwig files from hg19 version to hg38 version. And I asked chatGPT to understand the chain file downloaded from UCSC.

PPS:

If you want to learn Bioinformatics, there are three 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!