BIRS Workshop Lecture Videos

Banff International Research Station Logo

BIRS Workshop Lecture Videos

Surveys of Subtle Allelic Imbalance in Tissue Scheet, Paul


Somatically-acquired allelic imbalance (AI) is an established factor in cancer initiation and has recently been implicated as a marker for cancer risk. While DNA microarrays and next-generation sequencing are effective for whole-genome profiling of AI, in typical settings their sensitivities become extremely limited when the aberrant cell fraction (or tumor purity) is below 10-20%. Yet, this range may be critical for early detection and diagnostics, since often for such applications the samples of interest will be comprised of heterogeneous mixtures of cells with a large component of DNA from normal (i.e. the germline) rather than aberrant (e.g. the tumor) sources. Here we introduce a powerful haplotype-based computational technique (Vattathil & Scheet, 2013, Gen Res) and use it to characterize AI in several difficult settings. We start with a reanalysis of a study of over 35,000 samples of healthy tissue from recent genome-wide association studies and find a 2-fold higher rate of somatic mosaicism (within-individual genomic heterogeneity), which may indicate a wider applicability for the use of mosaicism as a biomarker for cancer risk. We next examine premalignant tissue, profiling polyps from individuals at risk for colorectal cancer to show subtle levels of AI across critical loci; we also demonstrate extensive mosaicism in the lung field (normal-appearing tissue surrounding the tumor), consistent with recent studies of expression (Kadara et. al., 2014, JNCI). Finally, we study lymph node tissue (of lung cancer patients), sampled via endobronchial ultrasound, and discover chromosomal aberrations in samples that were deemed negative by pathology review but that were ultimately determined to be positive following surgical extraction, thus demonstrating potential for molecular diagnostics.

Item Media

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International