The Power of “Lookup Biology”

Mar 28, 2016 · 3 min read

Guest post by Marina Sirota, PhD, twoXAR Advisor and Assistant Professor, UCSF Institute for Computational Health Sciences

Earlier this month, Andrew A. Radin and I had the opportunity to attend a community outreach meeting at UC Irvine hosted by the NIH Libraries of Cellular Signatures (LINCS) consortium. It was a great and diverse community gathering of drug discovery researchers from academia, biopharma, startups, consulting companies and government funding agencies. For anyone interested in listening to the talks, some of them have been posted on YouTube.

The focus of day one was to review the progress to date on collaborations among researchers spearheading projects that are exploring the relationship between cell signaling events, transcription and phenotypes across various perturbations and cell lines. Day two was a check-in on the joint activities of the LINCS Data and Signature Generation Centers (DSGCs) and the Data Coordination and Integration Center (DCIC) and discussions on future areas of improvement. The meeting also included presentations from several NIH institutes, the Allen Brain Institute, IBM and Google.

The LINCS program was launched in 2010, following the earlier Connectivity Map effort, to catalogue, analyze, and integrate molecular and cellular signatures that result from drug or gene-induced perturbations. LINCS contains molecular measurements on perturbations from more than 10,000 small molecules, knockdown and over-expressed genes, across 1,000 landmark genes as well as a number of proteins. As Broad’s Todd Golub and others mentioned in their talks, the goal of LINCS is to enable a new era of “lookup” biology or science that allows us to query rich data sets with complex questions of how compounds and genetics effect disease.

For drug hunters like twoXAR, the LINCS data can help answer questions like: Does a small molecule exist that modulates a molecular program of interest or what is the complete set of cellular responses to a drug candidate of interest? We can further integrate this great resource into the twoXAR platform as one of the “data layers” to model the molecular effects of drugs and enable computational drug discovery.

For clinical researchers, LINCS help them investigate how patient molecular response is related to biological mechanism. They can also look at a cellular signature and ask how it could be used as a measure of clinical response or if off-target toxicity can be parsed from on-target mechanisms.

In my own research, I used the earlier Connectivity Map dataset in combination with publicly available disease data from the Gene Expression Omnibus (GEO) to identify new indications for existing drugs[i]. Using this approach we were able to predict and validate the use of an anticonvulsant drug topiramate as a new therapeutic for Inflammatory Bowel Disease[ii]. More recently we have also used a similar approach to predict the use of alpha-adrenergic agonists as potential therapeutic agents for dermatomyositis[iii].

As a public resource, the LINCS data is freely available to anyone and there are a number of tools that can be used to analyze them (including these from the Ma’ayan Lab at Mt. Sinai). The kinds of queries you can make include:

  • Given a signature of a disease, search for disease-modifying perturbations
  • Given a measured response to a particular perturbation, search for other small molecules or genetic perturbations that induce a similar response
  • Given a cell type, look up the range of cell states that can be induced by changes in the microenvironment
  • Given a change induced by the microenvironment look up drugs that can reverse that change
  • Given a drug and a cell type, look up time-courses for induced changes in mRNA, protein abundance protein modification and cell phenotype

Given the amount of data and tools that have been generated by the LINCS consortium, there is a fantastic opportunity to start using and mining this rich dataset to identify new therapeutic targets and enable faster and more specific efforts in drug discovery.

While LINCS is still a work in progress, it stands as a shining example of how government funded projects, supported by academic research institutions, can serve as an invaluable catalyst to create public-private partnerships that advance our understanding of disease and foster the development of new technologies and medicines.

For more information on LINCS please visit:

[i] [ii] [iii]


Written by


Artificial intelligence-driven biopharmaceutical company. @softbank, @a16z, & @StartX backed.