Research Goal

We are particularly interested in elucidating tumor evolutionary dynamics, novel therapeutic targets, and the genotype to phenotype map in cancer. A unifying theme of our research is to exploit ‘omic’ data derived from clinically annotated samples in robust computational frameworks coupled with iterative experimental validation in order to advance our understanding of cancer systems biology. In particular, we employ advanced genomic techniques, computational and mathematical modeling, and elegant model systems in order to:

1) Model the evolutionary dynamics of tumor progression and therapeutic resistance

2) Elucidate disease etiology and novel molecular targets through integrative analyses of high-throughput omic data

3) Develop techniques for the systems-level interpretation of genotype-phenotype associations in cancer

Selected Publications

Between-region genetic divergence reflects the mode and tempo of tumor evolution, Nature Genetics 2017. [link]

A population genetics perspective on the determinants of intra-tumor heterogeneity, Biochimica et Biophysica Acta (BBA) - Reviews on Cancer 2017. [link]

Integrated genomic characterization of oesophageal carcinoma, Nature 2017. [link]

A Big Bang Model of human colorectal tumor growth. Nature Genetics, 2015. [link]

** Recommended by Faculty of 1000 Prime

- Nature Genetics News and Views Commentary: Big Bang and Context Driven Collapse [link]

Nature Reviews Cancer Research Highlight: Explosive Beginnings [link]

Contributions to drug resistance in glioblastoma derived from malignant cells in the sub-ependymal zone. Cancer Res 75(1), 2015. [pubmed]

Genome-driven integrated classification of breast cancer validated in over 7,500 samples. Genome Biology 15(8), 2014. [pubmed]

    Comprehensive Molecular Characterization of Gastric Adenocarcinoma. Nature, 2014, 513, 202-209. [pubmed]

    The breast cancer oncogene EMSY represses transcription of anti-metastatic microRNA miR-31. Molecular Cell, 53(5):806-18, 2014. [pubmed]

    The shaping and functional consequence of the miRNA landscape in breast cancer. Nature, 497(7449):378-82, 2013. [pubmed]

    Intratumor heterogeneity in human glioblastoma reflects cancer evolutionary dynamics. PNAS, 110(10):4009-4014, 2013 [Epub 2013 Feb 14]. [pubmed]

    Single-molecule genomic data delineate patient-specific tumor profiles and cancer stem cell organization. Cancer Research, 73(1): 41-49, 2013 [Epub 2012 Oct 22]. [pubmed]

    The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature, 486(7403):346-52, 2012. [pubmed]

Overview of the simulation framework and the genomic data analysis pipeline.

Given the implications of tumor dynamics for precision medicine, there is a need to systematically characterize the mode of evolution across diverse solid tumor types. In particular, methods to infer the role of natural selection within established human tumors are lacking. read more...

Figure 1. The Big Bang model of tumor growth.

(a) After initiation, a tumor grows predominantly as a single expansion populated by numerous heterogeneous subclones. ITH results from private alterations (colored arrowheads) that continuously accumulate owing to replication errors. In addition to public alterations present in the first transformed cell, private alterations acquired early persist and become pervasive in the final tumor although remaining non-dominant (colored segments). Late-arising alterations are only present in small regions of the tumor.

(b) In the Big Bang model, the pervasiveness of private alterations depends on when the alteration occurs during growth, rather than on selection for that alteration. The schematic illustrates how early private alterations, despite remaining non-dominant, are pervasive within the tumor (for example, red and yellow) and can be found in distant regions, thus appearing variegated (for example, red). This is owing to aberrant subclone mixing in the primordial tumor, followed by scattering during expansion. Late alterations are restricted to small regions (for example, black, pink, gray) and are essentially undetectable by conventional bulk genomic profiling. Distance from the dashed vertical axis corresponds to increasingly late onset for alterations. Dashed boxes represent sampled regions.

(c) We sampled an average of 23 individual tumor glands (<10,000 cells) from distant regions (~0.5 cm3 in size) and bulk (left and right) samples. Samples were profiled using several genomic techniques, including copy number analysis, whole-exome and targeted sequencing, neutral methylation tag sequencing and FISH, providing a panoramic view of genomic alterations throughout the tumor on multiple spatial scales.