The next steps for .NET Bio

Coordinator
Jul 23, 2012 at 1:08 PM

I'd like to start a thread that is a little presumptuous, and which might have a few hints of central planning. That isn't the idea of course, but in the universities we do have bright students looking for research and development projects, and there are enthusiasts around like the regulars on this forum who are actively looking to develop the libraries. 

So, building on the "what do want .net bio to be" thread, is there a particular area in which we should try to foster some development effort over the next 3-6 months? are there features that active members of the community are prepared to lead and describe them to the point that bright students might be drawn in as contributors? 

Some of the discussions to date suggest possibilities, but let me just put this out there and see what people say. Areas in which people are prepared to contribute ideas and actual code are the clearest indication of the right niche for the library. 

SO if you had a few days and some keen volunteers to do a .NET Bio hackathon, what would you do? 

cheers

jh 

Developer
Jul 25, 2012 at 11:20 PM

Just saw this project and I have this looks amazing!  I can think of a great many things that would be very useful.

On the more purely-research/academic side for projects (but with an applied focus), I think creating tools that would work with RAD tag sequencing data would be amazingly useful and mesh well with Microsoft Research expertise.  There are a number of issues that come up with RAD tag sequencing and a dearth of methods currently available.  How do you define clusters or call SNPs/variants?  How can one use that data in an integrative framework to make statements about the level of divergence between populations or areas of the genome that are unusual relative to the others?   How can you visualize and make intelligent statements about data quality and filtering levels?  I would be very happy to both contribute ideas, code and data for such a project.

On the more purely user-interface/integration end, I think building a front end for SNP calling and variant discovery from resequencing projects would be a HUGE asset to the field.  It looks like .NET bio contains the backend for a number of useful tools, and having an open source alternative to something like CLC Genomics Workbench would be really nice.  As a first pass, having a user interface for simply taking resequencing data from a known-organism and calling variants and other useful statistics would be very nice.  Such tools are out there, but I think the .NET framework could really add a lot.  I think with some good programmers doing a hackathon a lot could be done in a weekend.

-Nigel 

Coordinator
Jul 30, 2012 at 2:14 PM

This is a great post. welcome nigel and please feel free to flesh this out a bit more. NGS has been a major driver for some people using .NET Bio - some in the commercial space and some in academia - but the focus has been more immediately concerned with getting from reads to assembly. RAD tag approaches are not a settled space and so this is a very nice opportunity, and indeed one that links nicely to some work by other groups. 

A couple of high level questions: 

* As the area has been around a few years, is there a standard set of tools and workflows that people use? 
* Does this community have a wish-list that could be prioritised into chunks that people might undertake?

There are at least a couple of NGS people lurking on these forums and I know some of them have interests in the visualisation area. And of course everyone who has been involved in .NET Bio is keen to see applications that exploit the back end capabilities. But it would also be nice in an Agile way to build something basic, but immediately useful and grow around it. 

Please take this further below, and others please jump in. 

cheers

jh

 

Coordinator
Jul 30, 2012 at 2:21 PM

For those wanting background, try something like: 

https://www.wiki.ed.ac.uk/display/RADSequencing/Home

with software: 

http://creskolab.uoregon.edu/stacks/

with RADTools also on the page above. 

Aug 21, 2012 at 9:05 PM

This is great question. I'm sorry it has taken me so long to respond, but q late response can revive a valuable thread with new readers so here goes...

It would be great if the community could record ideas for contributions that would be appropriate for students in the issue tracker. In my experience there are always students (including life-long learners) looking for useful tasks to help them get to grips with a new tool. Furthermore, activities such as the Google Summer of Code require a project to have such a list of projects defined. It would be good to be prepared for next year.

I don't know how the Codeplex issue tracker would best handle this. In other tools I've used for this purpose we've added a "bite size" or "mentored" label so that people can easily search for such items.

In terms of what I (as a .net Bio observer rather than a knowledgeable bio-informatition) would like to see:

- tutorials on how to perform textbook analysis examples
- cool visualisation examples
- tests, test and more tests
- connectors to new open data sets

You also mentioned a hackathon - if the community can come up with enough good two day tasks for students would there be sufficient momentum and volunteer energy to make such a thing happen?

Ross

On Jul 23, 2012 7:34 PM, "jamesmhogan" <notifications@codeplex.com> wrote:

From: jamesmhogan

I'd like to start a thread that is a little presumptuous, and which might have a few hints of central planning. That isn't the idea of course, but in the universities we do have bright students looking for research and development projects, and there are enthusiasts around like the regulars on this forum who are actively looking to develop the libraries.

So, building on the "what do want .net bio to be" thread, is there a particular area in which we should try to foster some development effort over the next 3-6 months? are there features that active members of the community are prepared to lead and describe them to the point that bright students might be drawn in as contributors?

Some of the discussions to date suggest possibilities, but let me just put this out there and see what people say. Areas in which people are prepared to contribute ideas and actual code are the clearest indication of the right niche for the library.

SO if you had a few days and some keen volunteers to do a .NET Bio hackathon, what would you do?

cheers

jh

Read the full discussion online.

To add a post to this discussion, reply to this email (bio@discussions.codeplex.com)

To start a new discussion for this project, email bio@discussions.codeplex.com

You are receiving this email because you subscribed to this discussion on CodePlex. You can unsubscribe or change your settings on codePlex.com.

Please note: Images and attachments will be removed from emails. Any posts to this discussion will also be available online at codeplex.com

Coordinator
Sep 6, 2012 at 2:20 PM

Nigel, 

Did you have any further thoughts on the workbench idea? 

If there were to be a hackathon, might one around NGS garner some attention and participation? 

Other ideas? 

 

jh 

Sep 15, 2012 at 4:04 AM

I think that the best idea for a .NET Bio hackathon is looking a solution for padena util, to make this tool at level of ABySS, MIRA, and the biggests tools for assembly first generation and NGS. I'm using . Net Bio since MBF edition the last year, i want to become a .Net Bio guru for latin america and spain. please count with me.

Greeting from manizales, colombia

Leonardo

 

Developer
Jan 15, 2013 at 4:29 PM

Sorry for starting to contribute on this thread and then going dead for awhile, my dissertation wound up eating away my life for several months and I am just now getting back to .NET Bio.

 

In terms of things that would be great though, I think having a RAD package might be a bit too ambitious right now.  I am working on SNP calls, etc. so will try to start committing work as I write it back to the project.

Coordinator
Jan 15, 2013 at 11:55 PM

Nice! More contributions are always welcome.

As we discussed on a separate thread (actually comments on one of the bugs) and I'll repeat here for general visibility, anyone can contribute code. If your contribution is a patch that fixes a bug or makes some other change that does not add significant functionality, you don't need to sign any contribution agreement. If a contribution is a large amount of code and/or adds new features, this open source project needs to be sure that the work is yours and you have agreed to your work being distributed under the Apache 2.0 license. In that case we'll arrange for Outercurve (the nonprofit open source foundation that .NET Bio belongs to) to send you a contribution agreement which you would need to sign before we can accept your work.

Regardless, if any contributor sends us a patch, someone from the community will review it to make sure it works and complies with our coding standards, etc., and then patch the source tree for everyone to see and use your fix. The next time we do a new release (soon!) your fix will be available in the binary distribution as well.

I'm happy to answer any questions about the contribution process, and you can also find it detailed in the 'Documentation' section of this site.

Jan 16, 2013 at 12:40 PM

Due to my work on BWA, samtools etc. I started to use Github and quite like it. Now Windows Azure SDK team was using Github as well. I wonder if we could move from TFS? After getting used to Git I no longer willing to touch my legacy tfs stuff. No wonder people say Git is the second most important work by Linus after Linux kernel.

Best,

dong

Developer
Jan 16, 2013 at 3:12 PM

Thanks for the comments on getting contributions in!  I just sent a message to rick who can hopefully help me out with getting the patch setup.

Coordinator
Jan 17, 2013 at 12:42 PM

simon has covered someof the mechanics, but great to have you back, nigel et al.  dissertations have a habit of eating people's lives. i see the converstaion has moved on to anothrr thread, so i'll attack that tomorrow after i've looked at the links. best jh