**190**wrote:

I would like to check the correlation between two genes based on RNA-Seq data. I know that `rho`

is the Spearmans correlation coefficient.

Correlation coeffiecinet between -1 and 0 indicates negative correlation.

Correlation coeffiecinet between 0 and 1 indicates positive correlation.

When can you tell that the correlation is so strong? When I checked the correlation between two genes I see that `rho`

is 0.2. Can I say now that the two genes have a strong positive correlation? Is there any specific value to say that it is strong?

One more basic question: For having a look at correlation between two genes I used RNAseq data. Do I need to take only tumor samples or should I also consider normal samples for checking the correlation?

thanks

**30k**• written 20 months ago by Biologist •

**190**

I think there is more detail needed here about how the data is generated, but taking the statements at face value:

If the correlations (as you correctly stated) are in the interval

`[-1, 1]`

with positive and negative corresponding to the 'direction' of correlation, then I would say the answer is no you cannot say`0.2`

is strongly correlated. You can say it is positively correlated, but`0.2`

is much closer to`0`

than it is to`1`

.For arguments sake, a

strongpositive correlation would perhaps be`> 0.5`

. Similarly, a strong negative correlation might be`< -0.5`

. Thestrengthof the correlation is directly measured by themagnitudeof the number. That is to say, a correlation of 0 means the variables are entirely uncorrelated (obviously).For the last question, I would say the answer depends on the hypothesis you're aiming to test.

16kHere is a good explanation of why these rules of thumb do more harm than good: https://doi.org/10.1111/j.1467-9639.2009.00387.x

47kVery true. I was having a discussion earlier today about the arbitrary-ness of Fold Change cutoffs and how there isn't a one size fits all rule. Same logic applies.

Its all relative (which is kinda what I was alluding to by saying that we need more detail about this particularl experiment).

16k