Maybe, this two paper could help you (especially the first):
Fertil, B., Massin, M., Lespinats, S., Devic, C., Dumee, P. et Giron, A. (2005) “GENSTYLE: exploration and analysis of DNA sequences with genomic signature.” Nucleic Acids Res, Juil 2005 1;33 (Web Server issue):ws12-15.
2 Dufraigne, C., Fertil, B., Lespinats, S., Giron, A. et Deschavanne, P. (2005) “Detection and characterization of horizontal transfers in prokaryotes using genomic signature.” Nucleic Acids Res, Jan 2005 13, 33(1):e6
Why? If the original data was not collected in 2 base encoding, there is nothing at all to be gained by pseudo-converting a base space file to color space. And there is no way to truly convert a single base encoded dataset into a 2 base encoded data set - the best you could do is some sort of pseudo-conversion, but it will never be true "color space" data.
The documentation for cutadapt says that it will do any necessary conversion itself (https://cutadapt.readthedocs.org/en/latest/colorspace.html#ambiguity-in-colorspace)
e.g. from the manual "In colorspace mode, the adapter sequences given to the -a, -b and -g options can be given both as colors or as nucleotides. If given as nucleotides, they will automatically be converted to colorspace."