I have noticed that the beta spin channel in COHP calculation is multiplied by (-1) when it is compared with other codes and literature. In particular, I am focused on bulk Fe (conventional cell) whose input is as follows:

NEWK
16 32
1 0
COHP
1 500 19 30 1 12 0
-1 1
-1 2
END

In the pdf attached, I’ve represented the COHP for bulk Fe with two codes, Crystal17 and Siesta. In Crystal, the results have been taken directly from the COHP.DAT file. As can be seen, the shape of the alpha spin channel (blue) is quite similar in the two plots whereas the beta contribution (red) seems to be multiply by (-1) with respect to Siesta results. I have also found this change of sign by comparing with the results of R. Dronskowski, G. A. Landrum Angew. Chem. Int. Ed. 2000, 39, 1560-1585.