Bioz Launches Life Sciences Search Engine

By Allison Proffitt

July 20, 2016 | Bioz, Inc. came out of stealth today with the introduction of a search engine for life science experimentation. The company, started by Stanford research scientist Karin Lachmi and serial entrepreneur Daniel Levitt, also announced $3 million in seed funding, led by 5AM Ventures. Stanford-StartX Fund, Astia Angels and other institutional and individual investors, including Esther Dyson also invested in the round.

Each year, researchers in academia and biopharma spend $80 billion to purchase millions of products (reagents, consumables and instruments) for use in life science experiments, but there is no source for feedback on those purchases, Lachmi told Bio-IT World. “It’s easier today to book a hotel or buy shoes online than it is to find information I need for my research.”

Lachmi brought her idea to Levitt who recruited two software engineers with experience in natural language processing. Together, the four built Bioz (www.bioz.com).

The Bioz platform uses natural language processing and machine learning to extract data from published research articles on which products were used, under what conditions, for what experiments, and with which companion products. A Bioz algorithm ranks products based on how many times it is used in experiments, the impact factor of the journal in which papers are published, and how recently a product has been used. The result is Bioz Star rankings, which are meant to be objective measures of a products usefulness, not user-generated reviews.

Bioz “provides an indication of likelihood [that a product will] work successfully in your experiment,” Levitt says. “Researchers are super happy to have a way—which wasn’t available until today—to make better choices.”

For the most part, Bioz isn’t recording negative reviews. All of its data is gathered from the published literature, so it is recording products that worked. But each of the 200 million products in the database will have a product page, and researchers are welcome to comment on particular products through a community feature

Use Case

It was important to Lachmi that Bioz be built to align with how researchers actually work, “not only to choose the right product from the right vendor, but also to know if [a product is appropriate] for the experiment they are trying to do,” she said.

Researchers start by searching for any type of product by name, by category, or by assay/experiment (e.g. "antibodies", "cells", "kits", "PCR", "sequencer", etc.), Levitt explained. Bioz includes every product used in experimentation: biologicals, chemicals, glassware, pipettes, test tubes, and even instrumentation including sequencers, PCR machines, freezers, and centrifuges.

Bioz is more than just product listings; it includes structured information about the protocol conditions listed in the publications. That’s extremely useful context for researchers, Lachmi said. “Ok, I know what experiment I‘m trying to do and I know the product, but now I want to go further,” she suggested. “Should I use it at room temperature? Should I use a 1:1,000 dilution or a 1:200 dilution?”

Bioz is free for researchers to use, and the products within the database are limited only by publication. If a researcher wants to buy a product for an experiment, a button within the Bioz listing directs her to the product’s vendor. “Upon clicking on that button, the vendor is paying Bioz for the lead referral,” Levitt explained.

He sees expansion opportunities for the model in the future. Eventually Levitt hopes to collect a portion of the sale by delivering higher sales along with leads and customer metrics. Levitt didn’t rule out the possibility of Bioz becoming a marketplace itself eventually, but that’s not a model that the company is interested in pursuing now. Bioz has approached about 40 vendor companies so far, Levitt said, “and 100% of them have been interested in joining our program.”

The company reports over 30,000 beta users already from academic research labs and industry R&D labs, from more than 1,000 universities and biopharma companies, from 40 countries.