Is there any good statistic way or tools to compare the read count

[owwa]

Is there any good statistic way or tools to compare the read count difference between treatments?

The data is per-base sequencing read count, and we don’t know the exact distribution of the per-base read count (attached).

For example:
2 treatments (treatment1 & treatment2), 3 samples (biological repeat) in each treatment.

data table:
chrom position treatment1_1 treatment1_2 treatment1_3 treatment2_1 treatment2_2 treatment2_3
chr12 26229353 1 245 175 0 0 1
chr14 60833709 120 0 81 0 0 1
chr2 87516663 1 111 83 0 1 0
chr14 60802415 101 0 61 0 0 1
chr11 60356985 17 37 24 0 0 1
chr1 18300478 14 34 24 1 0 0
chr1 271501486 29 28 12 1 0 0
chr15 75932698 32 25 11 0 0 1
chr1 188326280 14 24 28 1 0 0
chr7 43038815 16 37 11 1 0 0
chr20 48971845 23 18 22 0 0 1
chr13 50030361 10 37 16 0 1 0
chr5 58346829 9 35 18 1 0 0
chr14 60803076 74 0 107 0 0 0
chr15 21554556 33 44 19 0 0 0
chr21 8881378 30 51 14 0 0 0
chrX 56348140 0 51 40 0 0 0
chr24 22387291 30 43 16 0 0 0


The aim is to know if there is significant difference between treatment1 and treatment2, a p value will be needed. Of course, if a FDR is also available, that will be fantastic.

Many thanks!

[man4ish]

How can i post a question on this forum?

How can i post a question on this forum?

[dpryan]

@owwa: One of the normal RNAseq programs (DESeq2, edgeR, etc.) will likely work. They expect reads counts like your and use a negative binomial model, which is likely what you need anyway.

@man4ish: If you click on one of the forums you'll see a "New Thread" button on the left side above the posts.

[owwa]

@owwa: One of the normal RNAseq programs (DESeq2, edgeR, etc.) will likely work. They expect reads counts like your and use a negative binomial model, which is likely what you need anyway.

@man4ish: If you click on one of the forums you'll see a "New Thread" button on the left side above the posts.

Many thanks to dpryan!

But my data is per-base read count, not like the RNAseq one. It is possible that the distribution of these two types of read count is different. I don't know if there is some tools or programs which are specific to per-base data.

[rnaeye]

How can i post a question on this forum?

Hi!
It's unfortunately an annoying page design:
Click on forums,
Choose a topic,
At that page, at top-left (or around there), you will see a box says "New Thread". That is how you post a new question.

[owwa]

Hi!
It's unfortunately an annoying page design:
Click on forums,
Choose a topic,
At that page, at top-left (or around there), you will see a box says "New Thread". That is how you post a new question.

yes,
1). go to: http://seqanswers.com/forums/index.php

2). select a topic, such as "Bioinformatics"

3). click the button "New Thread" (top-left of the page)

4). fill in your question title, content..., then submit.

[dpryan]

Many thanks to dpryan!

But my data is per-base read count, not like the RNAseq one. It is possible that the distribution of these two types of read count is different. I don't know if there is some tools or programs which are specific to per-base data.

Yes, it's possible that your case follows a different distribution, but likely not. The only way to find out is too look at the data and try to see what distribution it follows.

[owwa]

Thank you dpryan, your suggestions is very helpful.