a code for htseq analysis

Question

[Learner]

Is there anyone who has done htseq-count analysis and can share his script? the purpose is to find up and down regulated genes.

[WouterDeCoster]

This question shows you put minimal effort into trying to solve this yourself.

The manual of htseq-count is quite extensive and you should be able to figure this out on your own. If you get stuck you can still ask a more specific question.

I would suggest featureCounts instead of htseq-counts. It's faster and does almost the same. Both tools do not identify up and down-regulated genes. They only do read counting. For further analysis, you need a tool such as DESeq2 or edgeR.

[Learner]

@WouterDeCoster I appreciate the time you spent but I am wondering why you reply me always in a matter that I cannot follow. I totally understand that the english might not be your first language, however, I wrote above this "htseq-count analysis and can share his script? the purpose is to find up and down regulated genes." I want to find a way to end up to up and down regulated genes. Is this possible with htseq-count or I should look at another files ? if it is possible, which types of normalisation is used? why DESeq2 and not Limma ?( if one normalised the count values)

[WouterDeCoster]

why you reply me always in a matter that I cannot follow

I'm not sure which part of my question is unclear. If you can ask a more specific question I can elaborate.

I totally understand that the english might not be your first language,

That's right, but that's not the problem here.

I want to find a way to end up to up and down regulated genes. Is this possible with htseq-count or I should look at another files ? if it is possible, which types of normalisation is used?

htseq-counts is a good start. It will count reads, which you can then use in a program for differential expression analysis (see below). But a better choice is featureCounts, because it is faster and scales across cores. Another alternative would be Salmon, but that's not entirely the same. I just mention it here for completeness sake.

if it is possible, which types of normalisation is used?

You don't have to do any upfront normalization. Your software for differential expression analysis (see below) will take care of that.

why DESeq2 and not Limma ?( if one normalised the count values)

As far as I know, the most commonly used and recommended tools are DESeq2 and edgeR. You can use limma-voom, but not the standard limma. The standard limma is for microarray data, which follows a different statistical distribution.