Question: Finding most co-expressed genes of input gene based on Pearson correlation
1
gravatar for kozhaki.seq
4.3 years ago by
kozhaki.seq50
Korea, Republic Of
kozhaki.seq50 wrote:

I have a matrix of microarray (around 30000 genes X 800 samples) as a back-end data set. In the workflow, if someone input a gene of their interest, I need to find it's highly co-expressed genes based on back-end data set. For that, I need to calculate pairwise Pearson correlation of input gene against data set (in this case, 30000 pairs). I have tried psych package before and for this pairwise calculation, there might be some other better method. Also,I assume this process take a little long time. Can any experienced person can suggest on this?Thanks..

ADD COMMENTlink modified 4.3 years ago by Zhilong Jia1.5k • written 4.3 years ago by kozhaki.seq50

So you want a faster method for calculating the correlation matrix? Although you don't need WGCNA, I do believe the people in WGCNA has a faster implementation for the calculation of correlation. you can have a look

ADD REPLYlink written 4.3 years ago by Sam2.5k

Yes, since my matrix dimension is high, I am concern about the time..Yes, I will look into it @Sam

ADD REPLYlink written 4.3 years ago by kozhaki.seq50
2
gravatar for Zhilong Jia
4.3 years ago by
Zhilong Jia1.5k
London
Zhilong Jia1.5k wrote:

amap::Dist, a R function, can calculate the distance or correlation in parallel. See manual http://www.inside-r.org/packages/cran/amap/docs/Dist

ADD COMMENTlink written 4.3 years ago by Zhilong Jia1.5k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 866 users visited in the last hour