Hi Gang,
Thanks for your help. That makes a lot of sense.
One more thing...
Given the way that I have constructed my interaction terms using 1deval, should I think of the signs of the z-transformed correlation coefficients as an indication of how each interaction term is correlating with the different weights that comprise the interaction term and not in terms of correlation and anticorrelation?
For example, if the following 1deval call is for my '5nR' bar:
1deval -a Deconv5_HR_First_LftNAcc_v2.1D -b Deconv5_HR_Str1_LftNAcc_v2.1D -c Deconv5_HR_Str2_LftNAcc_v2.1D -d Deconv5_HR_Str3_LftNAcc_v2.1D -e Deconv5_HR_Str4_LftNAcc_v2.1D -f Deconv5_HR_Str5_LftNAcc_v2.1D -g Deconv5_HR_Ref_LftNAcc_v2.1D -expr '(-1*a)-(b)-(c)-(d)-(e)+(2*f)+(3*g)' >Deconv5_HR_Inter_LftNacc_Str5nRef.1D
Does the negative '5nR' bar in the previously attached file mean that that particular region is correlating more with the average of a, b, c, d, and e?
OR
Does the negative sign imply that that region is anticorrelated with the seed region during contexts Str5 and Ref?
If the signs are not indicative of correlation and anticorrelation, is it appropriate to think of the previously attached graph showing increasing and decreasing correlation with the seed region?
Thanks,
Aaron
