As the price of diagnostic and sequencing applied sciences has fallen quickly in recent times, researchers are gathering unprecedented quantities of information about illness and biology. Sadly, scientists who need to develop new therapies from knowledge typically want the assistance of somebody with software program engineering expertise.
Right this moment, Watershed Bio helps scientists and bioinformaticians run experiments and acquire insights with a platform that enables customers to research advanced datasets no matter their computational abilities. The cloud-based platform offers workflow templates and customizable interfaces that permit customers to discover and share all varieties of knowledge, together with complete genome sequencing, transcriptomics, proteomics, metabolomics, high-content imaging, protein folding, and extra.
“Scientists need to be taught concerning the software program and knowledge science a part of the sector, however they do not need to be software program engineers who write code simply to grasp knowledge,” says co-founder and CEO Jonathan Wang ’13, SM ’15. “With Watershed, you do not have to try this.”
Watershed is utilized by massive and small analysis groups in trade and academia to speed up discovery and decision-making. As new superior analytical methods are revealed in scientific journals, they are often immediately added to Watershed’s platform as templates, making cutting-edge instruments extra accessible and collaborative for researchers from all backgrounds.
“Organic knowledge is rising exponentially, and the sequencing applied sciences that generate this knowledge are getting higher and cheaper,” Wang says. “This downside coming from MIT was proper in my wheelhouse. It is a tough technical downside. It is also a significant downside as a result of these individuals are engaged on curing illnesses. They know there’s worth in all this knowledge, however they’re having a tough time leveraging it. We need to assist them unlock extra insights, quicker.”
code not detected
Wang had deliberate to main in biology at MIT, however he rapidly turned enthusiastic about the opportunity of utilizing laptop science to construct options for tens of millions of individuals. He finally earned each a bachelor’s diploma and a grasp’s diploma within the Division of Electrical Engineering and Pc Science (EECS). Wang additionally interned in a biology lab on the Massachusetts Institute of Know-how, the place he was stunned by how gradual and labor-intensive the experiments have been.
“I see the distinction between biology and laptop science, the place there’s a dynamic atmosphere. [in computer science] “You may get fast suggestions, and even in the event you’re the one particular person writing the code, there’s lots of issues you may work with immediately,” Wang says.
Whereas engaged on machine studying and high-performance computing at MIT, Wang co-founded a high-frequency buying and selling firm with a number of classmates. His crew employed researchers with PhDs in fields reminiscent of arithmetic and physics to develop new buying and selling methods, however quickly found bottlenecks within the course of.
“Researchers have been used to constructing prototypes, so issues moved slowly,” Wang stated. “These have been small approximations of fashions that may very well be run regionally on a machine. Bringing these approaches into manufacturing required engineers to run them in a high-throughput means on a compute cluster. However the engineers did not perceive the character of the analysis, so there was lots of backwards and forwards. Concepts that we thought may very well be applied in a day took weeks.”
To resolve this downside, Wang’s crew developed a software program layer that makes constructing production-ready fashions as straightforward as constructing a prototype on a laptop computer. Then, a couple of years after graduating from MIT, Wang realized that applied sciences like DNA sequencing have been changing into cheaper and extra widespread.
“The bottleneck wasn’t sequencing anymore, so individuals stated, ‘Let’s sequence every thing,'” Wang recollects. “The limiting issue turned computation. Individuals did not know what to do with all the info being generated. Biologists have been ready for knowledge scientists and bioinformaticians to assist them, however these individuals did not essentially perceive biology at a deep sufficient stage.”
The scenario appeared acquainted to Mr. Wang.
“This was precisely what we noticed within the monetary trade: researchers tried to work with engineers, however the engineers by no means absolutely understood, and there have been individuals ready for the engineers, which was inefficient,” Wang says. “However, we discovered that biologists are hungry to do these experiments, however there’s an enormous hole they usually really feel like they should grow to be software program engineers or simply give attention to science.”
Wang formally based Watershed in 2019 with Mark Kalinich ’13, a former MIT classmate and doctor who’s not concerned within the firm’s day-to-day operations.
Since then, Wang has heard from executives at biotech and pharmaceutical corporations concerning the rising complexity of organic analysis. Deriving new insights will more and more require evaluation of information from genome-wide, inhabitants research, RNA-seq, mass spectrometry, and extra. Growing customized therapies or choosing affected person populations for medical analysis can require huge datasets, and new methods to research knowledge are continuously being revealed in scientific journals.
Now, companies can run large-scale analytics with Watershed with out organising their very own servers or cloud computing accounts. Researchers can work quicker with ready-made templates that may deal with all the most typical knowledge sorts. Well-liked AI-based instruments like AlphaFold and Geneformer are additionally accessible, and Watershed’s platform makes it straightforward to share your workflows and dig deeper into your outcomes.
“This platform hits the candy spot of ease of use and customization for individuals of all backgrounds,” Wang stated. “In science, no two issues are precisely the identical. I keep away from the phrase product as a result of it means deploying one thing and operating it at scale without end. Analysis is just not like that. Analysis is about developing with an thought, testing it, and utilizing the outcomes to give you different concepts. The quicker you may design, implement, and run an experiment, the quicker you may transfer on to the following experiment.”
Biology accelerates
Wang believes Watershed helps biologists sustain with the newest advances in biology and accelerating scientific discoveries within the course of.
“If scientists can get insights a bit bit quicker, 10 or 20 occasions quicker, that may make a giant distinction,” Wang stated.
Watershed is utilized by researchers in academia and companies of all sizes. Biotech and pharmaceutical executives additionally use Watershed to make selections about new experiments and drug candidates.
“We have had success in all of those areas. The widespread thread is that individuals perceive analysis, however they don’t seem to be laptop science or software program engineering consultants,” Wang says. “It has been lots of enjoyable watching this trade develop. For me, it is nice to graduate from MIT and are available again to Kendall Sq., the house of Watershed. That is the place so most of the cutting-edge advances are occurring, and we’re making an attempt to do our half to allow the way forward for biology.”
