Welcome to the Exchange

From environmental samples, to cell-lines to small-molecule libraries, and even ‘omics’ data, researchers collect and create billions of dollars’ worth of research products each year. Most of this inventory, however, remains locked up in freezers and in storage, mostly because of the way research information is disseminated, with a near-total reliance on publications alone, and the absence of channels where scientists can exchange research collections. For instance, outside of published (positive) results and some centralized repositories, it is impossible to estimate the number of unique cell lines that exist and can serve as valuable models for cancer research, many of which were developed through NSF/NIH funded grants. Similarly, many researchers approach the National Park Service each year for environmental sample collection permits, but once collected, such samples disappear into lab freezers forever. The field of archeology appears to be undergoing a “curation crisis”, where entire collections, acquired at a considerable cost, remain un-curated and are eventually forgotten. This suggests that the lack of any downstream channels to share products and information not only makes research studies more difficult, it may also lead to grant dollars being spent on developing the same or similar products year after year.

​

We are addressing this significant unmet need in the sciences through Sample Exchange (SE), a new service and platform that will allow individual researchers, organizations and companies to share, exchange, loan, license and sell research collections and intermediate process products. SE will aggregate services and represent a new ecosystem for science, playing a role similar to Amazon, eBay, Alibaba and Craigslist combined. SE will also be a big step towards decentralizing repositories and creating a more distributed, stable market that is better able to serve research needs, while also being vastly cheaper to maintain and support than large, individual repositories.

​

Additionally, through SE, we intend to create an ultra-modern database architecture supported by deep learning, to specifically allow unique searches on sample attributes in multidimensional space. The database and platform, when fully developed, will be capable of complex searches for features based on natural language queries using GPT-3 (or similar) frameworks. Most importantly, the architecture will facilitate knowledge discovery by aggregating data in ways not possible when relying on keyword searches alone.

 

Seed funding for SE will help us reach the following goals:

1. Develop the SE ecosystem to allow material exchange, create new marketplaces and new avenues to collaborate

2. Spur new markets for product intermediates, allowing studies to start from intermediate points, reducing research waste while saving time and money

3. Create secure, trackable chains of exchanges, especially with regard to high security and BSL-rated samples.

4. Introduce the next generation of databases and searches that rely on natural language processing and not on keywords alone.