GENOMETRY, INC. is a life-science technology company dedicated to L1000 Expression Profiling, a revolutionary high-throughput low-cost gene-expression profiling method. L1000 produces high-dimensional gene-expression profiles direct from cells, at a rate of thousands a day, and at only a fraction of the cost of conventional methods. This enables the use of expression profiling at a scale never before possible. GENOMETRY is the sole and exclusive provider of commercial gene-expression profiling services using L1000, and provides data-generation and involved expert support to the pharmaceutical, biotechnology, personal care, consumer products, cosmetics, specialty chemical, and petrochemical industries from headquarters in Kendall Square, Cambridge MA.
Full-genome expression profiles can be faithfully recreated from the levels of just 5% of the transcripts. Measuring only these special landmark genes makes high-dimensional gene-expression profiling faster and cheaper than ever before.
L1000 Expression Profiling measures 978 validated landmark genes in crude cell lysate in 384-well plate format then calculates the levels of the remaining transcripts using an algorithm trained on tens-of-thousands of historical gene-expression profiles.
Library characterization, primary screening, hit prioritization, lead selection, novelty biasing, MOA elucidation, target deconvolution, safety alerts, indication discovery, and direct and seamless comparison with public L1000 data from the LINCS Program.
L1000 Expression Profiling is a patented whole-genome gene-expression assay based on the direct measurement of a reduced representation of the transcriptome and computational inference of the portion of the transcriptome not explicitly measured. The number of landmark transcripts whose abundance is measured is 978. An additional 80 invariant genes are also explicitly measured to enable scaling and normalization. The input for L1000 is crude lysates of human cells in 384-well plates. The output is tab-delimited text files of log-transformed expression values for 22,000 genes × 380 samples.
While working at the Broad Institute of MIT and Harvard University, GENOMETRY’s founders discovered that by measuring the expression levels of approximately one thousand specially-selected genes and using a computational model trained on more than one hundred thousand microarray experiments to infer the activity of the other genes, it was possible to produce close approximations of gene-expression profiles in which the levels of all twenty thousand transcripts are explicitly measured.
Our team then developed a multiplex ligation-mediated amplification assay to monitor landmark-gene expression and collaborated with Luminex Corporation in the creation of its FLEXMAP 3D® system, an instrument specifically designed to quantitate these analytes quickly and cost-effectively in 384-well plate format. The resulting L1000 Expression Profiling method delivers 80% of the information of a whole-genome assay for about 5% of the cost, with a throughput of tens-of-thousands of samples a week. Data produced using L1000 has the same format and is on the same scale as conventional microarray data.
L1000 has been extensively validated at the individual gene and full profile level, for both technical and operational performance, in hundreds of diseased and normal cell types. The technology has been in continuous operation at GENOMETRY since 2012, and at the Broad Institute since 2010 where it is being used to create a public reference collection of one million gene-expression profiles under the NIH-funded LINCS Program.
GENOMETRY is the exclusive licensee of patents covering L1000 owned by the Broad Institute, the Massachusetts Institute of Technology and the Dana-Farber Cancer Institute, and has rights* to additional intellectual property required to practice the assay commercially. The minimum order is 12 × 384-well plates (4,560 samples). Turnaround time is 2 weeks. Volume discounts are available for library-scale projects.
*This service is sold pursuant to licensing arrangements with Life Technologies Corporation and Luminex Corporation.
GENOMETRY is pioneering a digital open innovation platform in partnership with pharmaceutical companies and other owners of proprietary chemistry. By coupling our unique library-digitization capability with a Google-like search engine, GENOMETRY engages the global biomedical-research community in the drug-discovery enterprise. Individual researchers find potential chemical solutions for their chosen biological problems and professional drug-development partners. The owners of those compounds find new uses for existing chemical assets and committed domain-expert collaborators to help prosecute those projects.
Connectivity Map is a collection of gene-expression profiles from cultured human cells treated with hundreds of known drugs coupled with a dedicated pattern-matching data-mining algorithm, created by GENOMETRY’s founders while at the Broad Institute. It was designed to test the principle of using gene-expression profiling as a universal functional bioassay, and to demonstrate that matching diseases with small molecules through the transitive property of common or reciprocal gene-expression changes was a viable drug-discovery strategy (Lamb et al., Science 2006; Lamb, Nature Reviews Cancer 2007).
Gene-Expression Based Small-Molecule Screening The disease of interest is represented by the list of genes found to be aberrantly expressed in tissue specimens or a faithful model system. A library of small molecules is screened for compounds increasing the levels of down-regulated genes, and decreasing the levels of up-regulated genes.
Connectivity Map also demonstrated the power of digitizing, distributing, and democratizing the drug-screening process. The small-molecule gene-expression profiles and the search algorithm were made freely available on the internet, and found the global biomedical research community with an enormous appetite to mine these data. Thousands of scientists with world-leading disease-specific expertise, unique secondary assay systems, rare patient populations, and novel therapeutic concepts, but no access to small-molecule libraries or screening capabilities, now had a way to screen for drugs.
This approach has been described as “Google for drug discovery.” This is because just like typing a word or phrase into a traditional internet search engine and being presented with a list of websites relevant to that input, users of Connectivity Map upload a list of genes characterizing their disease of interest, and with one click see a list of small molecules ranked by their affect on those genes. These compounds are potential modulators of the disease.
The self-service discovery paradigm has now been applied successfully in dozens of published cases, and has led to the identification of agents with in vivo proof-of-concept against biologies as diverse as skeletal muscle atrophy (Kunkel et al., Cell Metabolism 2011), hair loss (Ishimatsu-Tsuji et al., FASEB J 2009), inflammatory bowel disease (Dudley et al., Science Translational Medicine 2011), glucocorticoid resistance (Wei et al., Cancer Cell 2006), osteogenesis (Brum et al., PNAS 2015), and obesity (Liu et al., Cell 2015).
GENOMETRY is now creating a new and improved version of Connectivity Map called Pollen. The most important difference is that Pollen is populated with novel, proprietary, drug-like chemistry from the libraries of major pharmaceutical companies. This means that companies with the skills and resources to translate discoveries into therapies have an immediate vested interest in the connnections being made. All of the contributors are therefore commited to make their compounds available to Pollen users to allow those users to test their hypotheses and document their discoveries in their own labs. GENOMETRY acts as the honest broker between users and the owners of the small molecules to establish the agreements required to pursue and co-develop promising connections.
Use of Pollen is free and unlimited.
Merck Uses L1000 Data from Genometry to Deorphanize “Dark Chemical Matter”
Boston MA, United States—27 February 2017—A paper published in Drug Discovery Today describes the strategies used by Merck Research Laboratories to understand and exploit so-called Dark Chemical Matter (DCM). DCM is the subset of a small-molecule library that has never shown any activity despite being tested in hundreds of assays. However, rather than simply being inert, these compounds may instead be exquisitely selective for unappreciated bioactivities, and thus highly desirable as novel leads or as tools with which to find new drug targets. Two complementary high-throughput linker-free technologies—Automated Ligand Identification System (ALIS) and L1000 Expression Profiling—have both the performance attributes and the broad and unbiased biological coverage required for efficient deorphanization of DCM, and are Merck’s methods-of-choice for this emerging discovery opportunity.
Merck Applies Innovative “Deep Learning” Methods to L1000 Data from Genometry
West Point PA, United States—9 February 2017—Scientists from Merck Research Laboratories today reported the application of deep learning techniques to the analysis and interpretation of L1000 Expression Profiling data. Using an illustrative set of 3,699 compounds from multiple active discovery campaigns the Merck researchers demonstrate that their new methods reveal chemical structure, protein target and promiscuity information encoded in the data, and outperform existing approaches for assigning functions to unknown compounds. The computational efficiency of the methods reported is of particular significance as gene-expression profiling becomes an increasingly important tool for small-molecule characterization, and the size and scope of the associated datasets get ever larger. The work is published in PLOS Computational Biology.
Expression-Based Drug Screening in 3D Cultures Uniquely Enabled by Genometry
Uppsala, Sweden—27 October 2016—Scientists from the Department of Medical Sciences at Uppsala University today published a paper entitled “Large-Scale Gene Expression Profiling Platform for Identification of Context-Dependent Drug Responses in Multicellular Tumor Spheroids” in the journal Cell Chemical Biology. The study used L1000 Expression Profiling to characterize the effects of small molecules at multiple doses on various types of multicellular spheroids in 384-well plate format, and revealed profound and therapeutically-relevant differences in transcriptional responses as a function of cellular disposition. It also demonstrated that L1000 is a powerful and practical universal assay technology for high-throughput screening in complex cellular systems.
Genometry Generates Data for Harvard Study Published in Nature Chemical Biology
Cambridge MA, United States—14 March 2016—Researchers from Massachusetts General Hospital and Harvard Medical School today published a paper entitled “Light-Controlled Modulation of Gene Expression by Chemical Optoepigenetic Probes” in the journal Nature Chemical Biology. L1000 Expression Profiling was used in the study to characterize the effects of a series of novel photochromic histone-deacetylase inhibitors across a large matrix of conditions, thereby demonstrating a powerful and generalizable new technique for chemical biology named Chemo-Optical Modulation of Epigenetically-regulated Transcription (COMET). Genometry scientists are co-authors of the paper.
Genometry Announces Deal with Janssen for Library-Scale Gene-Expression Profiling
Cambridge MA, United States—8 October 2015—GENOMETRY, INC. today announced entering into a service agreement with JANSSEN PHARMACEUTICA NV of Beerse, Belgium under the terms of which Genometry will generate gene-expression profiles from 250,000 compounds from Janssen’s small-molecule screening library using its proprietary L1000 Expression Profiling technology. This is the first time gene-expression profiling has been applied at this scale, and this multi-year contract showcases a commitment to high-throughput gene-expression profiling for multiple drug discovery and development applications. The data will be used for primary screening and library characterization as well as to improve the selection of candidate drugs prior to clinical studies. “L1000 was developed to make gene-expression profiling experiments of unprecedented scale and scope possible, and Genometry was founded to get this transformative technology into the hands of professional drug developers,” said Justin Lamb, Ph.D., Genometry’s President and CEO. “We are therefore thrilled to see Janssen engage in a true library-scale profiling effort uniquely enabled by Genometry’s capabilities.”