We have analyzed over 10 million open- and closed-access publications — a lot goes into whether we show a paper on the platform
For a figure from a publication to make it onto the BenchSci, we have to be highly confident in the data that our machine learning (ML) extracted from the figures and text. As part of the pipeline, our image recognition ML interprets figures to determine the type of experiment being shown and, if it expects there to be an antibody involved, it searches for a product citation in the paper. Ideally, the antibody is associated with a catalog number or a clone ID. In the case of a catalog number, we compare that to vendor catalogs already represented on BenchSci. However, we can still detect experiments when catalog numbers aren't cited, too. We have to be confident in the data at each step before the figure is shown on the platform.
Here are a few strategies to help you find a specific paper:
Check if the product you used is associated with your paper:
To do this, search by the catalog number of the antibody and check the figures on the product page. If there are many published figures available for the product, try sorting the results by “Publication Date” to easily find your publication.
If you did not cite a catalog number but you listed the vendor, start a search with the target protein/gene and then apply the company and application filters to home in on your publication. Sorting by Date or impact factor could also help you locate your paper.
If the paper is closed-access...
An updated list of our publishing partners can be found here. If your article is from a different closed-access publisher, we haven't gotten a hold of it quite yet (hold tight!).
Still couldn't find the paper?
We'd be happy to look into this for you, so please let us know using the in-app chat or by email at firstname.lastname@example.org.