Esther Rheinbay, PhD
Assistant Professor of Medicine
Massachusetts General Hospital Cancer Center
Harvard Medical School
Center for Cancer Research
Explore the Rheinbay Lab
Most known genomic drivers of cancer are in coding genes, affecting the encoded protein’s interaction with other proteins, DNA or biological compounds. Recent advances in DNA sequencing technology have made it possible to study non-coding regions that regulate these protein-coding genes. Several cancer drivers have been identified and characterized in these regulatory regions, however, this genomic territory remains relatively unexplored in human tumors. The Rheinbay laboratory concentrates on identifying and functionally characterizing these non-coding drivers in the sequences of tumor whole genomes through development of novel analysis strategies and collaborations with experimental investigators.
We are also interested in tumors, especially breast cancers, for which no known protein-coding driver alterations have been found. In the age of targeted therapy, these tumors pose a special challenge in that they leave few treatment options for patients beyond conventional chemotherapy. We believe that finding novel genomic and epigenomic, protein-coding and regulatory therapeutic targets in these tumors will have significant clinical implications.
Regulatory driver mutations in cancer genomes
Genomic cancer driver discovery has traditionally focused on protein-coding genes (the human exome), and large-scale sequencing of these genes in thousands of tumors has led to the discovery of novel frequently altered genes. However, exome sequencing focused only on coding genes does not allow analysis of non-coding regions in the human genome. Protein-coding genes are regulated by several types of genomic elements that control their expression (promoters, distal enhancers and boundary elements), translation (5’UTRs) and mRNA stability (3’UTRs). Alterations in the DNA sequence of these elements thus directly affect the expression and regulation of the target gene. Several such non-coding elements have been identified as recurrently altered in human cancer, and functionally characterized, although these non-coding drivers appear infrequent compared to protein-coding oncogenes and tumor suppressors.
One reason might be that gene regulation is highly tissue specific, and therefore driver alterations in non-coding regions might create a fitness advantage in only a single tumor type. Finding such a specific driver requires a sufficient number of whole genomes from this tumor type. With recent advances in DNA sequencing technology and an increasing number of whole cancer genomes available for analysis, we are just starting to map out and characterize regulatory driver alterations. The Rheinbay laboratory works on the development of novel methods to identify non-coding driver candidates using genomic and epigenomic sources of information, and to understand their impact on tumor initiation, progression and treatment resistance through collaborations with experimental colleagues.
We have recently identified a recurrent mutation in the promoter of the breast cancer oncogene FOXA1. This mutation increases expression through augmenting a binding site for E2F, leading to E2F protein recruitment. In addition, FOXA1 overexpression leads to resistance to the breast cancer drug, fulvestrant. We are now investigating the implications and mechanism of action of this mutation in breast cancer progression and treatment resistance.
Finding targetable vulnerabilities in cancers without known drivers
From recent large genome and exome sequencing studies of different cancer types, it has become apparent that there are almost always patients whose tumors harbor no common driver alteration such as BRAF mutation in melanoma, HER2 amplification, or hormone receptor expression in breast and prostate cancer. In an era of treatments targeting such alterations specific to a patient’s cancer cells, a lack of potentially druggable cancer drivers severely limits the repertoire of available therapy options. Rather than being truly without any drivers, these tumors are likely driven by yet uncharacterized protein-coding or regulatory genomic alterations, or an oncogenic state induced and maintained by epigenetic changes.
Our research is focused on finding the drivers and vulnerabilities of these particular tumors by integrating genomics and epigenomics data, with the ultimate goal of connecting patients to effective targeted treatments.
Rheinbay, E.*, Nielsen, M.M.*, Abascal, F.* et al. Analyses of non-coding somatic drivers in 2,658 cancer whole genomes. Nature 578, 102–111 (2020).
Campbell, P.J., Getz, G., Korbel, J.O. et al. Pan-cancer analysis of whole genomes. Nature 578, 82–93 (2020).
Rheinbay E, Parasuraman P, Grimsby J, et al. Recurrent and functional regulatory mutations in breast cancer. Nature. 2017;547:55-60.
Suva ML*, Rheinbay E*, Gillespie SM, et al. Reconstructing and reprogramming the tumor-propagating potential of glioblastoma stem-like cells. Cell. 2014;157:580-94.
Rheinbay E*, Suva ML*, Gillespie SM, et al. An aberrant transcription factor network essential for Wnt signaling and stem cell maintenance in glioblastoma. Cell Reports. 2013;3:1567-79.
We're Hiring! Postdoctoral Position
Postdoctoral Position in the Rheinbay Laboratory
Open position in computational cancer genomics to study cancer drivers and gender differences in cancer.
We are looking for a self-motivated postdoctoral researcher with a strong background in computational science and experience with large data sets. The successful candidate will join an interdisciplinary team working on rigorous analysis of next generation sequencing data (DNA, RNA, chromatin) from tumor samples, and development of analysis tools that will be shared with the research community. This is a unique training opportunity with access to resources at the MGH Cancer Center, Harvard Medical School and the Broad Institute.
- PhD in computational biology, bioinformatics, computer science, physics or applied math
- Strong publication record, communication and writing skills, fluency in English
- Excellent programming skills in Python and R, experience with cloud computing environments preferred
- Ability to work together with multi-disciplinary teams consisting of physicians, experimental scientists, statisticians and software engineers
- Experience with NGS data processing and analysis is desired
Interested candidates should submit a cover letter, curriculum vitae, research background and interests and contact information for three references to: Dr. Esther Rheinbay, email@example.com.
We're Hiring! Data Analyst Position
Data Analyst - Cancer Computational Biology
Unique opportunity to join an interdisciplinary team bridging the Harvard Medical School, the Massachusetts General Hospital, and the Broad Institute of Harvard and MIT. The Rheinbay Lab and Lawrence Lab at the Massachusetts General Hospital Cancer Center seek well-qualified candidates to join a team of computational biologists working at the forefront of cancer research and treatment. We use computation as a powerful microscope to study both the fundamental biology of cancer initiation and progression, as well the diagnosis and treatment of cancer patients in the hospital setting.
Current research interests:
- Cancer driver genes: tumors grow because of specific driver mutations that deactivate tumor suppressors or activate oncogenes. We are working to complete our understanding of the full catalog of cancer's "box of tricks".
- Resistance to targeted therapies: single drugs targeting specific driver mutations can be effective for a while, but the cancer invariably discovers a work-around. We are actively investigating mechanisms of drug resistance and how to combat it.
- Single-cell sequencing: New approaches allow us to dissect a tumor down to single cells and investigate the RNA expression or DNA mutations in each cell. Understanding intratumoral heterogeneity is shedding new light on cancer progression and patient outcomes.
- Liquid biopsies: novel state-of-the-art technologies are starting to allow us to monitor the progression of cancer (both before, during, and after treatment) through a simple blood draw. We are actively working to overcome analytical challenges inherent in the study of circulating tumor cells (CTCs) and cell-free circulating tumor DNA (ctDNA).
- Mutational processes: our genomes accumulate mutations from environmental agents such as ultraviolet radiation and tobacco smoke, as well as from intrinsic processes like errors during DNA replication. Studying these mutational background patterns can tell us what repair pathways are broken in a specific tumor, perhaps pointing the way to an effective genotoxic therapy. We are working to develop novel DNA sequencing technologies for studying mutagenesis in model systems
- Genomics of sex chromosomes in cancer: understand the genetic underpinnings of different incidence and outcome between men and women patients
- Work with Cancer Center researchers to understand experimental procedures and the kinds of data produced (e.g. DNA sequencing, RNA sequencing, epigenetic readouts, clinical outcome annotations). Meet and discuss with clinical and experimental colleagues to identify analytical challenges and goals.
- Apply existing and novel algorithms to cancer data sets, analyze data quality, critically review and analyze results, communicate results to biologists, computational biologists, software engineers and clinicians.
- Explore novel data visualization tools, with emphasis on integrating diverse data types and extracting clinically relevant insights.
- Contribute to scientific writing and creation of data figures to be included in research publications reporting novel discoveries made in the lab and clinic.
- B.A/B.S. in one of Computational Biology, Bioinformatics, Biology, Computer Science, Mathematics, Physics, or a related quantitative discipline.
- Independent, self-motivated drive to push research forward.
- Excellent programming skills (using any of Matlab, R, Java, Python, Perl, C, etc.)
- Nimble approach to programming and data analysis, with an emphasis on simple, intuitive, reasoning: quickly open unfamiliar datasets, generate simple visualizations to project the data onto our brains as usefully as possible, to stimulate hypothesis generation and the next steps of the analysis.
- Comfort using Word, Excel, Powerpoint or Google Suite tools to communicate results between team members.
- Ability to work together with multi-disciplinary teams comprising physicians, biologists, statisticians, and software engineers.
- Strong organizational and record-keeping skills
- Fluency in spoken and written English
- Experience in "machine learning" welcomed
Hotspot mutation in the FOXA1 promoter in breast cancer and proposed mechanism of action.
Esther Rheinbay, PhDPrincipal Investigator
- Joanna McDonald
- Meifang Qi, PhD