Null hypothesis in DESeq2

0

Entering edit mode

6.1 years ago

bioinfo456 ▴ 150

Hi all,

Does DESeq2 involve the concept of null hypothesis? I'm assuming yes, because it uses p value to conclude whether a particular gene is significant or not. If so, what is the definition of null hypothesis in this context? Again, I think it is "gene is not significant", which is assumed to be true initially before calculation of p value after which it is decided true or false. Please share your thoughts regarding this.

null hypothesis DESeq2 • 2.1k views

ADD COMMENT • link 6.1 years ago by bioinfo456 ▴ 150

1

Entering edit mode

Yes it does use null hypothesis-base significance testing. X is not signficant is not a statistical hypothesis; depending on your experimental setup / design etc the null used by DESeq2 would typically be that "Coefficient Y is zero for gene X".

ADD REPLY • link 6.1 years ago by russhh 5.7k

0

Entering edit mode

Thanks. What does Coefficient Y correspond to in the context of finding significant genes? Significance itself?

ADD REPLY • link 6.1 years ago by bioinfo456 ▴ 150

1

Entering edit mode

What are you talking about? Significance is a technical term; a "significant gene" in the context of a hypothesis test for a given coefficient (or contrast) is one for which the probability of getting a more extreme observation under the conditions of the null hypothesis is less than your threshold.

ADD REPLY • link 6.1 years ago by russhh 5.7k

0

Entering edit mode

yup got it! can you please explain to me the relation between Wald's test and negative binomial generalized linear model? As for my understanding, the count data is modeled using negative binomial generalized linear model after which Wald's test is applied to figure out whether a particular gene is significant or not. Please correct me if I'm wrong.