Improvements in CLI & additional subcommands

[OndrejSladky]

• Improved CLI structure
• Improved output headers
• Added subcommands for conversion to/from SPSS and splitting/joining ms.
• Possibility to construct maxone mask during masked superstring construction.

[PavelVesely]

A small glitch in naming the commands: the ms command outputs a .msfa file with mask-cased superstring , whereas in commands ms2msfa and msfa2ms the MS abbrev refers to separate superstring and mask. Not sure how to make it better...

[OndrejSladky]

Yeah, I do agree with this, I could rename it to sth like compute which'd be more descriptive than msfa.

[OndrejSladky]

Also I'm not sure, if -M should not output the masked-cased superstring maximizing ones instead of the mask, as it feels kinda anoying to run msfa2ms and then ms2msfa to obtain it.

[PavelVesely]

Also I'm not sure, if -M should not output the masked-cased superstring maximizing ones instead of the mask, as it feels kinda anoying to run msfa2ms and then ms2msfa to obtain it.

Yes, I think -M should output a .msfa file as it can then be used in FMSI-memb directly (right, it'd be annoying to run to more commands to obtain it).

[karel-brinda]

Comments:

• agree with compute
• optimize is misleading - probably maskopt?
• ms2msfa - Split a masked superstring into a superstring and a mask. - should be the other way around
• ./kmercamel msfa2ms doesn't contain any message
• is there any better way how to call M+S separately? Maybe mssep? It would be nice to change ms->mssep, msfa-> ms so that the basic format is the most easy to write.

[karel-brinda]

Unless it's critical for the current submission, we can finalize it next week.

[PavelVesely]

I will try to use the new CLI and streaming in a new version of the experimental pipeline, hopefully this week. Some renaming is still fine

[karel-brinda]

@OndrejSladky Could you give us a rough timeline for this? So we know what kind of additional feedback you need from us at each stage.

By the way, I’m really looking forward to seeing this integrated! :)

[OndrejSladky]

@OndrejSladky Could you give us a rough timeline for this? So we know what kind of additional feedback you need from us at each stage.

By the way, I’m really looking forward to seeing this integrated! :)

I'll update the changes you've suggested, likely tomorrow, and then we can iterate other suggestions on how the commands should be named and similar, until we're all satisfied. Then I'll merge (likely including the streaming alg) and bump to v2.0.0 fixing the interface.

[OndrejSladky]

Comments:

* agree with `compute`

* `optimize` is misleading - probably `maskopt`?

* `ms2msfa    - Split a masked superstring into a superstring and a mask.` - should be the other way around

* `./kmercamel msfa2ms` doesn't contain any message

* is there any better way how to call M+S separately? Maybe mssep? It would be nice to change `ms`->`mssep`, `msfa`-> `ms` so that the basic format is the most easy to write.

Agree with all of them. mssep sounds good.

[OndrejSladky]

I renamed everything as discussed and also made -M parameter output the masked superstring and not only the mask.
Additionally, I tried to improve the header and comments of the fastas outputted.
If you could take a look if now it looks good for you, then I'll merge.