Question: Two questions about pathifier package
0
14 months ago by
arronar0
arronar0 wrote:

Hello out there.

I'm currently trying to use pathifier package against a microarray dataset. That specific dataset looks like :

 Genes Cell-line-A Cell-line-A-Treated Cell-line-B Cell-line-B-Treated Cell-line-C Cell-line-C-Treated Gene1 exprs.val exprs.val exprs.val exprs.val exprs.val exprs.val Gene2 exprs.val exprs.val exprs.val exprs.val exprs.val exprs.val

The total genes are 22,000.  Anyway, the point here is that the dataset has only one biological replication from each Cell-line as you can see and only one biological replication for each treatment.

Question #1 :

Currently as normals argument at quantify_pathways_deregulation() function I give a vector of TRUE , FALSE , TRUE , FALSE , TRUE ,FALSE . But this isn't actually the right approach because these three cell lines are a bit different between each other. So the second thought here is to split that dataset into 3 sub-datasets according to cell lines and use as normals argument a vector of TRUE, FALSE and run the function 3 different times for each cell line. What is your opinion about that?

Question #2:

As I was reading the documentation of the pathifier, in order to configure it properly for my dataset, I saw that for the min_std argument they suggest to use the technical noise.

"min_std: The minimal allowed standard deviation of each gene. Genes with a lower standard deviation are divided by min_std instead of their actual standard deviation. (Recommended: set min_std to be the technical noise)."

But how am I going to calculate it? At first run, I just used the 0.2254005 value from the example they provide with that package, but this might not be the right for my dataset.

Any idea or hint is welcome.

written 14 months ago by arronar0

Content
Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.