I am sequencing microRNAs from post-mortem brain samples and it seems RNU44 is not conserved in some conditions. I need microRNA normalizer(s) that is not affected by post-mortem effects
The normalizer must be consistent in all of your samples, I suggest to use an endougeous control that highly expressed miRNAs in human such as let-7 family, miR-30a-5p or miR-148a-5p. You should find out the expression level of miRNAs in your sample before you select the control. RNUs are not always cosistent, but they might good if you use same sample types with same conditions.
I prefer U6 too. You can also try miR-16, RNU48. Some people use a mean of B-actin and RNU44 as the internal control, but I am not very happy with this. Others also do spike control. However B-actin values does not fluctuate much in post-mortem samples.