Deciding order of encodings

[BradyAJohnston]

I'm currently working on Julia BCIF support and while the decoding process is all going smoothly, I was wondering about the encoding process. Is there a standard order for encodings to be applied? I can think of some obvious orders that would make the most sense, but if there was particular order that molstar uses then I'd prefer to stick to that.

[dsehnal]

You are right the encodings should follow a specific order. It is whatever the user (or algorithm) determines. Some examples:

• https://github.com/molstar/molstar/blob/master/src/mol-model/structure/export/categories/atom_site.ts#L119
• https://github.com/molstar/molstar/blob/master/src/servers/volume/server/query/encode.ts#L54
• https://github.com/molstar/molstar/blob/master/src/mol-io/common/binary-cif/classifier.ts
• https://github.com/molstar/ciftools-python/blob/main/tests/_encoding.py#L57

[BradyAJohnston]

Okay just so I am 100% clear there isn't anything defined in the spec for what order they should be applied or an 'optimal' order of encodings?

[dsehnal]

No, this is in full control of the program that does the encoding.

[BradyAJohnston]

Okay thanks for the extra info and the examples!