I'm curious, why don't we post open source sequences more often?

[Jarrad]

Hey Guys,

I see alot mentions to do with open science but I rarely see full open source sequences? Why is that?

Would DIYBio be served borrowing some concepts from the software world in terms of their Free & Open Source Software movement? as well as using version control (git?) to document the sequence history.

AFAIK the main reason is current approach is very much a bottom-up approach, where it's easier and cheaper to use various enzymes to chop and insert various segments until you get what you want, then to provide a protocol on how to repeat this process... but an alterantive would be to provide with a complete sequence first, then allow the "compiler" (lab monkey) to generate a protocol?

That way multiple people could use the same source code sequence, adapt and fork it, but ultimately use various different techniques/protocols (such as dna printing) to get to the end result.

Or is there some other reason I've overlooked?

[Koeng]

The reason is that people don't really say it's "open". Sequences are just shoved it databases that you have to find. 

A good start is here http://parts.igem.org/Main_Page

A lot of times people just know what they want for a sequence then look for a plasmid that has it. For example, I could say "I want a pUC ori, AmpR, CEN/ARS, leu2 auxotrophy, and lacZ MCS." I would know exactly what I want (pRS315, http://www.snapgene.com/resources/plasmid_files/yeast_plasmids/pRS315/). Then I would ask around colleagues for it.

There's not enough DIYbiologists to make a type of git really worth it. I know what most work on and if I needed to know one of their sequences I'd pop them an email or post here. Also for the git it's important to remember that unlike computing biology is extremely *context dependent*. One ribosomal binding site will give different expression per protein (ranging hundreds times of expression). If I wanted to express something, I wouldn't go to a registry like the parts registry, I'd go find it myself from a virus or go to the RBS calculator (https://salislab.net/software/)

What do you want? If you know send me an email anytime and I bet I could find it. Sharing the physical DNA is a whole other rant though.

-Koeng

[Levi Morris]

As someone who has worked on both sides of the fence (software
engineering and genetic engineering,) I think that the issue of
context-dependent system behavior will resolve as the field grows.
Right now it's still very much a nascent field (in terms of
"development".) I'd say that in a lot of ways computing IS very
context dependent, the difference is that it is a much more mature
field and when I set out to develop a piece of software the problems
of context are fairly easy to resolve by carefully selecting a set of
tools for the task at hand.

Bioengineering just isn't there yet - to me genetics is still
essentially writing context-specific machine code. As time progresses
I expect that we'll see something akin to an operating system or a
framework on which higher-level development can be completed without
having to think about every low-level detail of the system in
question.

Now as far as just documenting the history of a particular sequence,
I'm kind of shocked that git isn't already being used for this. That
parts.igem.org is page is a pretty painful user experience and their
API doesn't look much better (XML?!?!?). It would be nice to be able
to clone a git repo of all that data and manipulate it as I see fit
(but man, XML?!?)

[John Griessen]

On 06/17/2015 03:48 PM, Levi Morris wrote:
> as far as just documenting the history of a particular sequence,
> I'm kind of shocked that git isn't already being used for this. That
> parts.igem.org is page is a pretty painful user experience and their
> API doesn't look much better (XML?!?!?). It would be nice to be able
> to clone a git repo of all that data and manipulate it as I see fit
> (but man, XML?!?)

WE have anti-XML rant discussions on a list about electronics tools development too.
It's sooo bloaty! DNA has redundancy and repetition for protection, but XML has it "just for fun".

There seems to be an anti-coder bias in bio.

John
(not a coder -- yet...I do hardware, and some of that requires code -- scripts not apps.)

[Cathal Garvey]

:) If there's an anti-coder bias, I haven't seen it yet! If anything,
other coders are the only people with strong opinions about things like
XML vs JSON vs YAML..

--
Scientific Director, IndieBio Irish Programme
Now running in Cork, Ireland May->July
Learn more at indieb.io and follow along!
Twitter: @onetruecathal
Phone: +353876363185
miniLock: JjmYYngs7akLZUjkvFkuYdsZ3PyPHSZRBKNm6qTYKZfAM
peerio.com: cathalgarvey