VigyaanCD: Bio/Chemical Modeling Workbench on PC

PKA writes "Announcing the release of VigyaanCD (homepage, mirror1, mirror2). Vigyaan, a Knoppix customization, is an electronic workbench for computational biology and computational chemistry. The user can choose from more than 20 science applications. In addition to bioinformatics software, it provides GROMACS, TINKER for biomolecular modeling and Ghemical, MPQC, PSI3 for quantum chemistry calculations. VigyaanCD is a live Linux CD containing all the software required to boot the PC into workbench environment and is suitable both for beginners (12 demos and tutorials) and experts. Learning computational biology/chemistry has never been easier, so grab the free ISO image from several mirrors around the world and put your PC to work."

Comments

I'll be the first to admit...

[HotNeedleOfInquiry]

That I don't know anything about biomolecular modeling. OTH, do we really want anyone creating their very own designer drugs/poisons/catylists/etc?

Maybe someone could explain whether or not it's possible to take a model and actually make a few grams of it.

Re:I'll be the first to admit...

[linzeal]

As freeman dyson stated as a challange in a wired interview I'm too lazy to dig up, " We need a machine that can physically place and read molecules at a rate suitable for manufacturing and not just research." You could do it today with the right equipment, but having the software does not make you proficient in using it or understanding its implications.

Re:I'll be the first to admit...

[Bowling Moses]

"OTH, do we really want anyone creating their very own designer drugs/poisons/catylists/etc?"

Sure we do. New drugs come from somewhere. Some are found using rational design, or are based on an initial hit from rational design. I'm working on a project right now to try and inhibit an enzyme that appears to be critical for establishment of a long-term infection of a certain major pathogen. One way of doing that is by taking the structure of the enzyme and trying to model in a whole series of compounds to find ones that appear to bind well to the active site. Another would be to attempt to solve the enzyme:inhibitor structure by x-ray crystallography, but crystallization's not always straightforward. Often the two methods are used in tandem.

"Maybe someone could explain whether or not it's possible to take a model and actually make a few grams of it."

Once you've got a model compound that you're interested in trying out in the real world, you're going to have to figure out how to obtain it. Various chemical and pharmaceutical companies (sigma, fisher, roche, or with all the consolidations lately is it sigmafisherrocheetc?) have hundreds of thousands (or more) compounds available for sale. Or you might be unlucky and have to make it. In that case, someone might have synthesized the compound for a completely different purpose already, in which case you might use a program such as Scifinder Scholar to search the organic chemistry literature to see if it's already been made. This program's not something that everybody has, my entire university's license only allows one person from the whole university to be logged on at a given time. If it's there, you're in luck and you "just" have to follow their materials and methods. Of course, it still could be a bitch. I've got a very bright undergrad working on synthesizing a compound for me that I found this way. In three months he's managed to get a trace amount that's highly impure--and this is with him working in an organic chemistry prof's lab with all the equipment and support necessary. If you're unlucky and it hasn't been synthesized or isn't in the literature, then you've got to figure the synthesis out yourself--which considering what the organic chemists have already made will likely require a hell of a lot of experience and expertise in chemistry to pull off--not just somebody out there in a basement lab. A basement lab might be just fine for synthesizing a number of previously known poisons or drugs, of course.

Re:I'll be the first to admit...

[InadequateCamel]

As far as drug production, I say go for it, but note that this approach WILL NOT show you HOW to make the molecule. It's not a matter of figuring out what the compound should look like, it's a matter of testing it and making the damn thing in the first place. That's why pharmaceutical companies need so much money; they need the best chemists to figure out how to make all these exotic drugs without the use of poisonous reagents, and then they need to make sure the drugs work the way you want (they usually do not).

On that note, if you want to make a poison it's not just a matter of tossing molecules at a computer; you gotta know how your prospective poison is going to do the poisoning...never mind the testing and actually making it.

Attention Slashdot:

[Neil Blender]

Learning computational biology/chemistry has never been easier,

This may give you a limited working knowledge, but only formal education and real professional experience will provide you the depth of knowledge to become a competent bioinformatics programmer (or chem or physics or whatever.) Learning computational biology is not easy. People with formal education in biology can become ok to decent programmers. But it very rarely works the other way around. I have worked in life sciences for 15 years, the last 5 in bioinformatics and I have seen it time and again.

Re:Attention Slashdot:

[untermensch]

This may give you a limited working knowledge, but only formal education and real professional experience will provide you the depth of knowledge to become a competent bioinformatics programmer

Agreed. However, this seems like a good way to introduce the interested life scientist to Bioinformatics, or even Linux in general. Also, the CD appears to contain not just programming tools, but general purpose scientific applications which could be effectively used by life scientists even with no programming knowledge whatsoever.

Is this the start of a new trend .... I hope so!

[rskrishnan]

Hello /.ers,
Is Vigyaan the start of a new trend in building "customized distros + bundled apps" by _vertical_. Not to sound like Mr.Marketeer - but by vertical i mean a narrow segment of users who care about say a handful of _must_have_apps_, along with the general utility/productivity apps.
I can imagine that having a specialized distro for the EDA/PCB/ design guys - that had say gEDA, pcb and few other "must_have_apps". This would be a total godsend for say someone who does pcb work for a living and can get right to the meat of his/her work and not have to worry about this that and the other. Specially with gEDA being somewhat of a monster to setup - since it has diff "sub-project" modules which all need to sync up to have something usable.

Anyway, EDA is vertical that I can think of - there must surely be MANY others - say something for the finance dudes/dudettes, or more such fields/areas.

Thanks,
RSK.

Re:Is this the start of a new trend .... I hope so

[LWATCDR]

GIS might also be a an interesting vertical. The thing I am amused by is how this all seems like a step back in many ways. In the old days when you wanted to work on a taks like accounting or programming you just stuck in the right boot disk and started the computer.

Re:Is this the start of a new trend .... I hope so

[rskrishnan]

Well - booting from a floppy!! Wow! I can remember those days - running TurboC (TC/TCC) from "ramdisk" on a PC with 640KB!! Days where you could literally carry "your stuff" with you in a box of diskettes and maybe a couple of tapes.

In comparison I think the set of choices available today simply is mind blowing!! Do you want to boot from the net, floppy + big disk, just a disk, USB, firewire + variants, and in the last 6 months or so .... I saw booting from a "ram disk" (i.e. basically RAM + some interfacing logic + battery pack - HOLY COW!!). Not to mention the host of "flash ram/disk/whatever" that's available now (non-trivial $$ but it's still a choice).

I love the set of choices - but what i really like is the ability to very niftily setup a realistic work env in a flash .... and leave pretty much nothing behind when you leave.

I somehow seem to gravitate towards the ability to walk-in to a place talk to the guys, setup shop in 30 mins, work for a steady period and then just leave - leaving _nothing_ behind you! No boot partitions, no email files, no /tmp crap absolutely _nothing_.

Almost seems spooky - when you think back to the era of either Windows with mystery files/registry crap + who knows what else, or the "big iron" Unix boxes which took hours and hours to setup/configure and get productive on!

Now with a Knoppix plus "job/role specific software" - I'm in virtual heaven! This could really be a big deal for an IT shop or a contractor or a "coders on demand" outfit.

But maybe I'm making too much out of this .... :)

Thanks,
RSK.

Vigyaan: Why not have ready to eat

[Anonymous Coward]

I think Vigyaan (knowledge in case any one doesnt know) is a great move towards user level customization. Linux can be brought out to the masses to the extent that the non techie BioCHem guy can take one customised cd and voila he doesnt ahve to mess with installs and more. So that would be great to encourage MORE linux usage.
On a side note, I would want to see a Freeware 3-D modelling software as Solidworks is a wee bit(3000$) too costly for student use!
Enjoy