Principle component analysis PCA

Question

[phamh]

Hi,

I would like to run PCA on my expression data but I do not know anything about command line as it is not required for my research. Is there any way that I can run PCA without using command line?

I looked it up and it seems like SPSS can be used to run PCA, but can it do it with expression data?

Thank you.

[ATpoint]

If you have no experience at all I recommend to partner with someone who can assist you. You probably find some tools where you can interactively press buttons but you have to be sure that your data are properly normalized and log2-transformed before you plug them into any PCA tool. You will also need to select a subset of genes that are informative for a meaningful PCA, no rocket science for experienced people but not trivial if you have no experience. Do you have a core facility at your institute for a group with a bioinformatician? A simple PCA takes a couple of minutes, maybe you can ask for a favor offering a beer in return.

In what format are your data and what are the data (RNA-seq?), are they normalized and log2-ed?

[phamh]

They are RNA seq data and normalized.

Answer 1

[GenoMax]

See: PCAtools: everything Principal Components Analysis

[phamh]

I just took a look at the link you provided. It seems like the instructions from those links are for command line which I did say I knew nothing about. I still appreciate you taking your time to respond to my post.

[GenoMax]

If you need a point and click interface for NGS data analysis they you are going to be limited to using commercial software packages/online tools that will require a license purchase. Galaxy (which is free) may have some limited tools available (I don't know if PCA is one of them).

It may be best to start learning unix/command line skills now, if you are going to keep working in bioinformatics.

[swbarnes2]

We know that sometimes a PI will tell some lab-based person "Hey, you are good with excel, so go figure out how to do this" to someone who doesn't know a thing, and doesn't really want to learn more than how to just do this one thing.. And if that's you, we're sorry, because most of this stuff can't be easily done in a free Windows program (and most of the people here don't use such tools even if they existed). So it seems to me your lab has two choices...pay for software that does what you want like, maybe CLC (where, if you pay them, you'll get support from them), or someone has to take the time to learn how to use the free tools, which means learning how to use command lines. The offerings for "free" and "come with a pretty GUI and work in Windows" are thin. That said, I google "PCA in Excel" and it didn't come up empty, so I'd look into that.