This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Problem in analyzing Rice RNA-Seq data

Hi friends

I have project on Rice RNA-Seq, in this project i have only 1 replicate for each of 3 conditions.(N,M,S) I should to say, Number of mapped reads (using tophat2) for conditions are:

N= 14,640,080 
M= 14,764,894 
S= 14,935,162

And you can see "S" has more mapped reads( in contrast with other ~17000 and ~29000).

I analyzed with cuffdiff (with blind mode) which generates:

N_vs_M=30
N_vs_S=350
M_vs_S=319

I know N_vs_M=30 has a problem in gene count and not fine.

I can not find out the problem , i also used DESeq,DESeq2 and edgeR but they could not show any significant DE genes.

I literature review and found same project where done by BGI company, i could not gain any information about the analysis process. They produced these columns:

geneID
geneLength
A-Expression
B-Expression
A-RPKM
B-RPKM
log2 Ratio(B/A)
Up-Down-Regulation(B/A)
P-value
FDR

I have 2 questions:

  1. Which software can produce these output
  2. How can I generate P-value and FDR without replicate? (I mean software)

Thanks

tophat deseq htseq cufflinks deseq2
• 3,655 views • link
updated by Ram Ram • written by Whoknows Whoknows

Hello pcsam.2008!

It appears that your post has been cross-posted to another site: SEQanswers

This is typically not recommended as it runs the risk of annoying people in both communities.

• 0 views • link
written by Devon Ryan Devon Ryan

Do you have another idea about this paper "Transcriptomic Analysis of Rice (Oryza sativa) Developing Embryos Using the RNA-Seq Technique" , they have only 1 sample per condition and also said:

We applied the R package DEGseq to identify DEGs with the random sampling model based on the read count for each gene at different developmental stages [33]. We used ''FDR <=0.001 and the absolute value of log2Ratio >=1'' as the threshold for judging the significance of each gene expression difference. More stringent criteria with a smaller FDR and bigger fold-change value can be used to identify DEGs

What do you think about this issue? - I mean 1 sample and FDR 0.001 and FC=2

How is it possible?

Thanks

• 0 views • link
updated by Ram Ram • written by Whoknows Whoknows

This should probably be its own post.

Having said that, that paper should have been rejected.

• 0 views • link
written by Devon Ryan Devon Ryan

1 answer

You need to contact the institution/company/individual that gave you the analysis for details.

That being said if you have no replication across conditions then there is very little chance of being able to meaningfully compare these conditions.

• 0 views • link
updated by Ram Ram • written by Istvan Albert Istvan Albert

thanks Istvan

• 0 views • link
updated by Ram Ram • written by Whoknows Whoknows

Log in to answer this question.

More posts like this