WHy can't we use biopython > 1.67 h

[nbisweden]

WHy can't we use biopython > 1.67 h

[Juke34]

Early 2019 :). We have re-developed the core of the tool to make it faster and add parallelisation, but since then we haven't had time to implement and test the numerous parameters we have in version 1.
Btw I have look into detail about the issue with Biopython version >1.67, and seems finally really easy to fix. I will push a fix soon.

[Juke34]

At the time we started the project it was biopython 1.67. When a new version came out we realised there was an incompatibility due to the fact they have deprecated and removed sub_features.
We should update the code to make it working with version >= 1.68.

[hinz1]

Exactly 1.67? So I would have to downgrade my system to run your tool?

[Juke34]

You could give a try to version >1.68. If it works we can just exclude version 1.68.
If it doesn't work don't downgrade your system, use Conda instead.

[tseemann]

I agree conda envs are a solution to this, but 25 August 2016: Biopython 1.68 is >3 years old now. Who knows what other bugs have been fixed?

[Juke34]

I agree. For our defence we use it just a little, few core methods so it guess it should not be any big difference even after many years.
We can probably use more recent version but we haven't try. I will put it in my to do list.
Anyway with the version 2 (branch feature/speedup) we should overcome this limitation.

[hinz1]

Well, we got by by installing on a machine that was rather "clean", python-wise, so installation went through and everything worked. So, thank you for publishing your work. I'll keep in mind that conda would be another way. As much as I like Python as a language, the ecosystem really is a mess. ;)

[tseemann]

What is the ETA for version 2?

[tseemann]

Thanks @Juke34 !