status update

[gracezheng]

Hi everyone,

It was great to work with all of you on this project.

I just want to send a quick update on some thoughts:

• It seems that people are interested in continuing working on this. Can you let me know if you are, and how much time you'll have in the next 2 months, so that we can figure out the project scope?
• Is F1000 the only place to publish this work? Any other options? Please contribute your ideas!
• Additional features:
  a) a more informative summary page: eg. something about the dataset, such as number of genes per cell, number of umi distribution per cell
  b) support >10k cells
  c) multiple samples?
• Testing:
  a) the apps should run on the web and off-line (if a user wants to run it on a laptop, the person should not have to install a bunch of libraries before being able to run it. We should probably take care of it.)
  b) make sure all the support features work well on multiple platforms.
  c) make sure all the logics make sense
  d) meaningful error messages and fail gracefully
• Documentation:
  a) help manual
  b) a brief tutorial to walk people through the app

10x is having a project planning meeting on Friday, and I will provide another update after that.

Thanks,
Grace

[oganm]

To contact everyone you probably want to send mails as not everyone are active github users.

I am interested though I am unsure about the amount of time I can assign. On weekends, I am available this weekend (29-30 Oct) then 26-27 November. I can arrange weekends after those as needed. Other than that I can put some hours irregularly on weekdays. That depends on how everything else goes.

For publication, I am unsure, I'll consult my lab to see what they think.

We probably want to decide what is the exact end-product will be before implementing anything else so next weekend seems like a good time to talk.

[esmason]

I can work on it a few hours a week, usually I have more time on weekends. I can chat about it this weekend too,

[szhan]

I am more interested in integration with gene ontology. I would be interested in working on it on weekends mostly.

[szhan]

In terms of potential publication, maybe PLoS ONE? Possibly Bioinformatics App Note (novel enough)? Maybe IEEE BioVis (which recommends papers to be published in BMC Bioinformatics)?

[oganm]

Plos one is probably not a step up from F1k so it might not worth it to discard the discount for it. BMC Bioinformatics would be nice though we might not make it there. Maybe as a first attempt. What is Bioinformatics App Note? I couldn't find anything about it

[szhan]

I meant Application Note in Bioinformatics (which is probably a long shot anyway). If we can't make it to BMC, then we can't make it to Bioinformatics... I am not too familiar with F1000, as I haven't really read any articles published there. There are also PeerJ and eLife, but I am not too familiar with them either.

[1234jc4321]

I found these two papers recently published in Bioinformatics as Application Notes. I think they are similar to VASCO. It could be a good place to publish.

Alla Mikheenko, Gleb Valin, Andrey Prjibelski, Vladislav Saveliev, and Alexey Gurevich
Icarus: visualizer for de novo assembly evaluation
Bioinformatics (2016) 32 (21): 3321-3323 doi:10.1093/bioinformatics/btw379

Hamid Younesy, Cydney B. Nielsen, Matthew C. Lorincz, Steven J.M. Jones, Mohammad M. Karimi, and Torsten Möller
ChAsE: chromatin analysis and exploration tool
Bioinformatics (2016) 32 (21): 3324-3326 doi:10.1093/bioinformatics/btw382

I'm also not familiar with F1000 research. The good point is that there is a hackathon channel and that one of the advisor of the channel is Ben Busby, one of the team leader at the Hackseq event.

I'm also interested to put a couple of hours per week in this project.

[oganm]

Any input from 10x? @gracezheng

[esmason]

What software quality does a visualization tool that is published in a peer reviewed journal have to be?

Because right now our code quality, robustness and maintainability is pretty low . Would we need to write unit tests, docstrings etc. ?

[oganm]

the main issue would be what exactly to put in the final product. then we can deal with making it maintainable. Though from my experience, code quality isn't a principle requirement for publication. It would make our lives better without a doubt though.