Choose Language Hide Translation Bar
Highlighted
Confa
Level I

Fit proportional Model for multiple genes/column at once

Hi
I would like to know if there is a way to make a script for returning risk ratios and p values for several columns/genes at once.

I mean I have a file with multiple columns (genes) and multiple row (patients). I have to analyze 100 genes for which i have the expression values categorized as UP- or DOWN- regulated and I want to get the Risk Ratios for all of them at once, without having to select a different gene each time in the dialog box. I use the script editor, but I am unable to automate this function for each column I want.

4 REPLIES 4
Highlighted
Confa
Level I

Re: Fit proportional Model for multiple genes/column at once

No Clue? Even with a script in R?

Highlighted

Re: Fit proportional Model for multiple genes/column at once

HI @Confa ,

 

Since the genes are responses, you would put all of them into the Y section of Fit Model. You do not have to put them one at a time. Select them all in the Select Columns section can drag and drop them into Y.

 

Is that what you were looking for?

 

Chris

Chris Kirchberg
Principal Systems Engineer, Life Sciences - JMP Global Technical Enablement
SAS Institute, Inc. - Denver, CO
Tel: +1-919-531-9927 ▪ Mobile: +1-303-378-7419 ▪ E-mail: chris.kirchberg@jmp.com
JMP – A Division of SAS Institute | www.jmp.com
Highlighted
Confa
Level I

Re: Fit proportional Model for multiple genes/column at once

Hi Chris, not really.
I have to calculate the Risk Ratios (Death and Time to Death) for more than 150 genes (upregulated vs downregulated), using the Fit Proportiona Hazard mode, not the Fit XY.

Here is the platform. Normally if you put one "gene" or variable in the box you will get the RR for that gene. If you put multiple genes or variable you will get a multivariable model, not the single RR for each separated variable.

fit.png

Highlighted

Re: Fit proportional Model for multiple genes/column at once

HI @Confa ,

 

Try Stacking all 150 genes. You will have two new columns. Something like Gene Name and Direction. Then use the By field to put the gene name in there and then the direction in teh Model Effects section. You will get a single fit for each gene.

 

Chris

Chris Kirchberg
Principal Systems Engineer, Life Sciences - JMP Global Technical Enablement
SAS Institute, Inc. - Denver, CO
Tel: +1-919-531-9927 ▪ Mobile: +1-303-378-7419 ▪ E-mail: chris.kirchberg@jmp.com
JMP – A Division of SAS Institute | www.jmp.com
Article Labels

    There are no labels assigned to this post.