LatchBio provides scientists with a no-code platform for biotech big data processing

- Advertisement -


Today, biologists and other scientists face a vast sea of ​​data and a staggering array of tools to process it, many of which require a specialist to work with. Hiring such a person is not an easy task, and the work can take months to complete … but LatchBio offers an option that allows you to process data through AlphaFold and other top-level tools in seconds. The company just raised $28 million to build its increasingly relevant platform.

- Advertisement -

Basically, the problem is that not all scientists are data scientists.

- Advertisement -

“Biologists, as great as they are at biology, pipetting, and lab work…they suck at programming,” said Alfredo Andere, co-founder and CEO of LatchBio. But the revolution in biotechnology is based on a huge increase in the data coming from each experiment.

Perhaps a few years ago it could be processed independently using basic tools, but the volume has increased a thousand times or more, and each new discovery (for example, cheaper genome sequencing or new ways to use this data) inflates it even more.

- Advertisement -

“If you are running a CRISPR experiment, after editing in the lab, you sequence it on the Illumina machine. It returns you a file with the RNA string and the changes you made, but it’s not one of them, but 10,000,” Andere continued. “If you are just a biologist, you need to learn how to use CRISPResso, use the command line, install dependencies, feed it the right data…”

The only real way to deal with this amount of raw data right now is to hand it over to a computer biologist, which is a rare breed and is rarely available on short notice.

“The problem we’ve seen over and over again is that these people are really hard to find — it’s like 20 biologists for one computational biologist. So they submit their data and wait, sometimes for months,” Andere said.

Andere and his co-founders, CTO Kenny Workman and COO Kyle Griffin, had strong backgrounds in technology but were disillusioned with the industry they worked in.

“We had amazing data pipelines at Google, but they were meant to serve ads. Then we saw how these biotech companies treat diseases, but they had the worst data pipelines in the world,” he said. So why not have the same power and ease of use as Google-level tools, but designed to be used by scientists who can’t write a single line of code? This is the goal of the company’s Latch platform, which is primarily focused on ease of use and flexibility.

“You really need to make it easy for biologists: you need tools that will allow them to upload data, then fill in about 3 parameters and click Run,” Andere said. “Like AlphaFold, this is a very heavy model. We saw it at Berkeley; they literally spent weeks trying to install it on a GPU cluster and failed. It’s so difficult. We gave them our platform, you inserted the amino acid sequence and launched it. We have [genetic sequencing pioneer] George Church himself the other day — literally in 30 seconds we got him to run AlphaFold on the platform.”

Launch AlphaFold on Latch in 4 easy steps: add, fill in a couple of fields, click “Run” and you’re done.

If all of this seems a little childish to scientists… ask. Biologists are likely to be the first to say they don’t want to deal with code. Good scientists are smart people, but they tend to want to focus on what they do best rather than learn a new discipline just to make sense of an avalanche of data. There are other scientists for whom this is a real job!

The problem arises when you consider how diverse the fields of biology and biotechnology are. Each field, such as proteomics, epigenetics, and dozens of subfields each, has many unique software tools and processes. Although some platforms and methods, such as Jupyter notebooks and the like, have become de facto standards in many fields, they are aimed at bioinformaticians rather than goggle-wearers in wet labs.

“Biology is so complex that you can’t have the generic tools we’ve built for software development,” Andere said. “So you have a chicken-and-egg situation: biologists won’t use this if there aren’t work processes, and there won’t be work processes if people aren’t using them.”

We got out of this situation, as he put it, by buying a chicken. They developed popular workflows themselves and passed them on to biologists, then used that feedback to improve their SDK so they could move into computational types with something easy to use. Andere noted that the alternative is often something like cloning GitHub repositories and other notebooks, or pulling code from documents and personal sites. Computational biologists dislike punishment more than their pipette cousins, so anything that makes it easier for them is welcome. Adding your process to Latch with the SDK means that scientists filling out their mailbox can do it themselves.

The hope and one of the main goals of this fundraiser is that the biotech community will continue to interact with the LatchBio platform, allowing the company to move from more mainstream workflows to making existing biotech infrastructure more accessible. Many companies have their own tools and stacks, but like everyone else, they often only work with experts. If that could change, it would free up these valuable data experts to create, not just implement.

“A company with 50,000 employees is not a customer right now because it has a hundred or 200 people doing computational biology. But if only these companies can build it, and smaller companies can’t, it’s an opportunity,” Andere said. “We have A and B series companies that can’t do it themselves and we work closely with them. We’re growing fast, and soon a D-series company starting to build its own will say, “Why do this when LatchBio works?” Companies will pay a lot not to build this house.”

The founders of LatchBio dance in dark clothing.

Founders of LatchBio in the lab.

The company expects ARR to be $1 million by the end of the year and contracts are piling up. But Andere stressed that the platform will always be free for scientists (even if their license situation can be painful), who tend to wait a month rather than spend five figures.

Round A, worth $28 million, was co-hosted by Coatue and Lux ​​Capital, with participation from Hummingbird Ventures, Caffeinated Capital, Haystack and Fifty Years.

Andere said they hope not just to create a product, but also to hire the best software development team in biotechnology. “I think young people are tired of working in optimization and quantification companies – there is no company in biotech that can claim to be one of the best software engineers in the world while working on world-changing problems.” he concluded. Naturally, LatchBio strives to be just that.


Credit: techcrunch.com /

- Advertisement -

Stay on top - Get the daily news in your inbox

DMCA / Correction Notice

Recent Articles

Related Stories

Stay on top - Get the daily news in your inbox