0
$\begingroup$

I have data on 2 histone marks that influence chromatin density. I have calculated individual correlation scores using the cor.test() function in R for both. I would like to combine the 2 together to get a combined correlation score but I'm not sure what the best method to do this is. Can I just add the data for the 2 together and get a score with cor.test()?

$\endgroup$
  • $\begingroup$ The histone data is percentage of Mb region covered by mark hotspot. I am also assuming each mark has equal effect on chromatin organisation. $\endgroup$ – Mattysi Jul 22 '18 at 14:08
0
$\begingroup$

the method of combining the two variables would be informed by some understanding of the markers, ie what is it that the combined measure is meant to characterise? maybe you want to take zscores and then sum or average them, at least that has appeared in the literature in other contexts, so perhaps there is a precedent to refer to

$\endgroup$
  • $\begingroup$ For each 1Mb section of the genome the percentage covered by chromatin mark hotspots has been calculated. The density per Mb has also been calculated. I want to see if an increase in chromatin marks correlates possitively or negatively with chromattin density. The density is affected by several different marks so I would hope the correlation would get stronger the more marks are included. $\endgroup$ – Mattysi Jul 22 '18 at 14:31
  • $\begingroup$ perhaps transform values to zscores and sum across the two variables then, that's not uncommon. I should hesitate to offer a suggestion because i'm not familiar with this type of data tho $\endgroup$ – pau13rown Jul 22 '18 at 14:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.