Science Exchange (YC S11): A Central Database For Core Research Facilities

http://techcrunch.com/2011/08/16/yc-funded-science-exchange-a-central-database-for-core-research-facilities/

by canistr • 15 years ago

136 26 15 years ago

lbarrow• 15 years ago

This is awesome! I'm really happy to see a start-up that contributes to scientific advancement in a meaningful, direct way. This service offers a straight-up efficiency improvement over the old method, in which most universities tried to replicate the core experimental facilities of everyone else.

recampbell• 15 years ago

It seems like this could be made into a even more generic experiment service. Sort of like Mechanical Turk for science experiments, where you pay other scientists to reproduce your results?

troymc• 15 years ago

I think perhaps you meant to write "...pay other scientists to repeat your experiments?"

(Their experiment might not reproduce your results. It happens!)

bbgm2• 15 years ago

There are other efforts to create exchanges or marketplaces for scientific research as well (e.g. http://www.assaydepot.com/). Small number, but definite possibilities. The problem is that a lot of research institutes and companies have a strong "we need to do everything" mindset that will take time to overcome

edit as noted elsewhere, funding models also implicitly encourage this behavior

hardtke• 15 years ago

The trouble with this idea is that university researchers have very little cash and are loathe to part with it. Almost all of their grant money goes to paying the salaries of people in their lab (fixed costs). If they need to use a piece of equipment at another institution, the current method is to propose a collaboration with a researcher at that university. Additionally, any decent researcher already knows where the specialized equipment resides (they can figure it out by reading the papers in their field or chatting at conferences), so there is no need for a centralized database.

elizabethiorns• 15 years ago

As an university researcher (faculty at University of Miami) I agree that efficiency of spending is very important - what is interesting is that researchers are often unaware that by making use of economies of scale that exist at specialized core facilities it is often cheaper than conducting experiments in house (a good example is cloning).

In some fields salaries might comprise the majority of lab costs but in biology this is not the case - experimental costs which can be spent as the researcher sees fit comprise the majority of costs. These costs can be outsourced to pay core facilities (currently $2B is spent outsourcing experiments to core facilities annually in the US).

While i agree that it is possible for researchers to establish where specialized equipment resides by reading papers and chatting with colleagues at conferences this is obviously inefficient and an incomplete record of potential providers. A centralized marketplace is a more efficient method for identifying providers.

tryitnow2• 15 years ago

"A centralized marketplace is a more efficient method for identifying providers."

That's spot on. Access to resources shouldn't be based on "special knowledge" Success in research should be a result of insights, creativity, and hard work, not "being in the know." Best of luck.

aorshan• 15 years ago

You work at the University of Miami? Any chance you have internships available for UM students? (I sent an email through the contact us page on your website as well)

jules• 15 years ago

> Almost all of their grant money goes to paying the salaries of people in their lab (fixed costs).

That highly depends on the field. At my university in the math department, you're completely right. In the physics department however, the costs of the machines and experimental setups dwarf the salaries. Because the demand for things is so low, the cost is ridiculously high. For example suppose there were no plastic coke bottles, but you require one for an experiment. How much do you think it costs to make one coke bottle in a university's workshop, or to order one from a specialized company? The costs can easily go in the thousands of dollars. That's the situation scientists are in for parts used in their setups, and for the machines they use. If you can share machines, that can be a huge cost savings.

eam2• 15 years ago

We actually launched a very similar website with the same goal earlier this year which is cool because it validates our idea. Works almost the same way! http://zombal.com/

equark• 15 years ago

Consider changing "Transform your scientific expertise into cash" to something like "Find consulting work given your scientific expertise."

chrislo• 15 years ago

Interesting. A lot of the "challenges" here: http://zombal.com/scientific-challenges seem like generic questions rather than requests for specific technical work/lab experiments (e.g. "Is dieting an ineffective way to lose weight?")

Is that a design choice? Is there as much overlap as you think between your site and the one mentioned in the original post?

eam2• 15 years ago

We actually have two formats, one is called Standard Zombs (standard tasks) which is bid-based just like ScienceExchange. The other is called Bounty Zombs (Q&A's) similar to http://prizes.org/ format. If you go to http://zombal.com/scientific-challenges as you pointed out you can filter out the different types.

tryitnow2• 15 years ago

It would be cool to see if amateurs/hobbyists could someday make use of services like this.

rorrr• 15 years ago

I really would love to see this work, but I have very very strong doubts. I hope you've done your market research, and aren't wasting time creating this website.

I've done scientific research, got published, so I saw this process first hand.

1) Grants are usually very small, only high end labs and a few science centers would do this (in my opinion).

2) Scientists are crazy about keeping their research private. There's a lot of competition and back stabbing going on in the scientific world.

3) Most scientists would find it very hard to rely on somebody they don't know. Experiments are extremely important, and often there's not much room for failure. Let's say you're a biologist who worked on a cell line for 3 years, and you finally have enough material to do an experiment. Would you ship it somewhere else to save some money?

4) Research projects that cannot be done locally are generally not approved. Meaning that the whole research project is financed and planned ahead of time. It's hard enough to get grant money, now imagine your proposed project has many more moving parts and points of failure.

bbgm2• 15 years ago

It will be very hard, given that the funding models don't necessarily align right now, but I think there is an opportunity. There's definitely an opportunity with small biotech, where there is need, but often no resources for specific project. FWIW in some areas, distributed research is becoming the norm just out of necessity.

hirenj2• 15 years ago

I wholeheartedly agree on points 2 and 3 - In fact, I've known scientists that were loathe to send BLAST jobs out to external servers.

Another point is that I've yet to see any core facilities that are running idle. Our core facility mass spec guys are running 24/7, and that's not even taking into account the time they lose to the instrument screwing up. In fact, it'll be hard to see some of the high end equipment running under heavier load – it's all just too fragile.

riprock• 15 years ago

I emailed my dad (biosci professor) for his opinion on this company, but from what I know of his work history and past collaborations, points 2 and 3 are likely road blocks to using this webapp. Trust is definitely a huge deal as I know he flies in specialized technicians he trusts over more "local" and likely "cheaper" technicians. Not saying the local technicians aren't trustworthy or aren't as skilled, but no matter the costs, trust founded on years of collaboration is irreplaceable.

That said, I'm guessing this kind of service may be more appealing to fresh PhDs.

brianbreslin• 15 years ago

I am pretty sure one of the founders is an actual research scientist. She came to this idea via a personal need. the other points i have no response to. - disclosure i'm friends with a few of the founders.

perlgeek• 15 years ago

I think there are two types of "experiments".

One of them is the classical scientific experiment, where you build an apparatus or somethings from dozens of components, and look at what results it produces. I can see how most of your concerns apply to that.

The other one is supplemental standard measurements, which can be also very expensive. For example if you create a new metal alloy, you might need an AFM image of the grain boundaries. If you receive such an image, you can very easily assess its quality, making it easier to rely on the "results" of others.

If you plan ahead, you can include a few thousand USD for AFM images in your grant application, much easier than getting a grant approved which includes buying your own AFM.

rdl• 15 years ago

I think this is a strong datapoint that YC isn't just random consumer web features!

hirenj2• 15 years ago

I just had a burl at the site, and threw in a couple of techniques that I would be interested in, or figured should be generic enough. The site seemed empty: Microarrays (error page), mass spectrometry, protein purification, NMR, PCR, biostatistics, bioinformatics.

Are there currently any open projects or facilities renting themselves out?

mrjasonroy• 15 years ago

I've worked in Scientific Websites (Biocompare the main one) for 6 years, and I love what Science Exchange is doing. The old guard of distributor networks, old boy research journals and closed research communities. It's taken a while but the industry is ready.

Biocompare received the same reaction when they launched - "scientists like their distributors and the free swag they get", "they'll never order products themselves". Research is ready for a disruption.

davidedicillo• 15 years ago

I wish I could see more startups like this one, that goes beyond the hot (and often useless) markets of the moment. Really smart people working to solve real problems or to optimize our resources. This is what this country need.

lurker14• 15 years ago

The world needs more YC startups like this.

anandkulkarni• 15 years ago

Brilliant. I can't wait to see how this expands to other kinds of scientific outsourcing - problems like scientific software development are highly specialized and surprisingly easy to outsource, but there's no good way for specialists to find folks interested in doing this kind of work.

Science Exchange (YC S11): A Central Database For Core Research Facilities

Related Stories