Genomic DNA Variation ComputerAided Discovery Methods Baylor College

Genomic DNA Variation Computer-Aided Discovery Methods Baylor College of Medicine course 311 -405 Term 3, 2010/2011 Lecture on Wednesday, February 2 nd, 2011 Aleksandar Milosavljevic, Ph. D http: //www. brl. bcm. tmc. edu

Cancer Genome Variation: Methods

Cancer Genome Variation: Sequencing Provides Comprehensive Tumor Characterization

Chromosome Aberrations: References 1 of 2 Background reviews [Balmain 2001] Balmain, A. , Cancer genetics: from Boveri and Mendel to microarrays. Nat Rev Cancer, 2001. 1(1): p. 77 -82. [Albertson et al. 2003] Albertson, D. G. , et al. , Chromosome aberrations in solid tumors. Nat Genet, 2003. 34(4): p. 369 -76. [Rabbitts et al. 2003] Rabbitts, T. H. and M. R. Stocks, Chromosomal translocation products engender new intracellular therapeutic technologies. Nat Med, 2003. 9(4): p. 383 -6. [Kumar-Sinha et al. 2008] Kumar-Sinha, C. , S. A. Tomlins, et al. (2008). "Recurrent gene fusions in prostate cancer. " Nat Rev Cancer 8(7): 497 -511. •

Chromosome Aberrations References 2 of 2 Research articles [Chin et al. 2006] Chin K. et al. Genomic and transcriptional aberrations linked to breast cancer pathophysiologies, Cancer Cell 10: 529 -541 2006 [Tomlins et al. 2005] Tomlins SA et al. , Recurrent fusion of TMPRSS 2 and ETS transcription factor genes in prostate cancer. Science, 2005. 310(5748): p. 644 -8. [TCGA 2008] The Cancer Genome Atlas Network "Comprehensive genomic characterization defines human glioblastoma genes and core pathways. " Nature 455(7216): 1061 -1068. [Miller et al. ] Miller C. A. et al. Discovering functional modules by identifyingrecurrent and mutually exclusive mutational patterns in tumors [in review].

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] single gene [Tomlins et al. 2005] none expression and other attributes of individual genes [Kumar-Sinha et al. 2008] networks / modules / gene sets background knowledge employed in discovery

Boveri, one century ago … Multiple cell poles cause unequal segregation of chromosomes. a | Fertilization of sea-urchin eggs by two sperm results in multiple cell poles. b | Chromosomes are aberrantly segregated [Balmain 2001]

Chromosomal aberrations [Albertson et al. ]

Chromosomal aberrations [Albertson et al. ]

Cancer Genome Variation: Methods

(Array) Comparative Genome Hybridization (array CGH)

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] single gene [Tomlins et al. 2005] none expression and other attributes of individual genes [Kumar-Sinha et al. 2008] networks / modules / gene sets background knowledge employed in discovery

Genomic and transcriptional aberrations linked to breast cancer pathophysiologies [Chin et al. 2006] • 100+ aggressively treated early stage breast tumors 1989 -1997, before ERBB 2 antagonist Trastuzumab (Herceptin) was approved for treating ERBB 2+ breast cancer

ERBB 2 heuristic (“paradigm”) formulated in last sentence of Chin K. et al. “Taking ERBB 2 as the paradigm (recurrently amplified, overexpressed, associated with outcome and with demonstrated functional importance in cancer) suggests FGFR 1, TACC 1, ADAM 9, IKBKB, PNMT, and GRB 7 as high-priority therapeutic targets in these regions of amplification. ”

“Taking ERBB 2 as the paradigm (recurrently amplified, overexpressed… Array CGH (~3 K BAC array) Gene expression (Affymetrix U 133 A array)

“Taking ERBB 2 as the paradigm (recurrently amplified…

“Taking ERBB 2 as the paradigm (… associated with outcome…)

“Taking ERBB 2 as the paradigm (… associated with outcome…)

Going beyond copy-number changes Deletions, amplifications induce aberrant fusions …but… Some aberrant fusion-producing rearrangements ( reciprocal translocations, inversions ) may not affect copy number

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] [Kumar-Sinha et al. 2008] [Tomlins et al. 2005] single gene none expression and other attributes of individual genes networks / modules / gene sets background knowledge employed in discovery

Two significant types of aberrant fusions aberrantly amplified expression aberrant activation of signaling protein [Rabbitts et al. ]

BCR-ABL fusion in Chronic Myeloid Leukaemia: four decades from lesion discovery to Imatinib ( Gleevec) 1960: Philadelphia chromosome discovered 1973: Chromosome translocation t(9; 22) identified 1983: Activated oncogene ABL identified 2001: Drug inhibiting BCR-ABL fusion identified

Fourfold significance of recurrent chromosomal aberrations Prognostic Marker Drug target Pointing to biological pathway Early diagnostic marker

Case Study: Prostate Cancer [Tomlins et al. 2005] Recurrent ( > 50% cases) chromosomal aberrations discovered in leukaemias, lymphomas, and sarcomas Carcinomas more complex: -- more rearrangements -- submicroscopic structure Gene overexpression recurrent chromosomal aberration present in > 50% prostate carcinomas [Tomlins et al. 2005]

Cancer Outlier Profile Analysis (COPA) using Oncomine database reveals overexpression of ETV 1 and ERG [Tomlins et al. ]

Frequent gene amplifications and losses in receptor tyrosine kinase-mediated signaling ETV 1 ERG

Recurrent TMPRSS 2: ETV 1 and TMPRSS 2: ERG fusions revealed by the study of rearrangements involving ETV 1 and ERG Expression of TMPRSS 2 is regulated by androgen [Tomlins et al. 2005]

Exclusivity of rearrangement: either ETV 1 or ERG [Tomlins et al. 2005]

TMPRSS 2 translocation associated with: • Aggressive disease Cancer Res 66: 8347 -51, 2006 • Reduced disease free survival Cancer Biol Ther 6, 2007 • Higher rate of prostate cancer specific death TMPRSS 2: ERG gene fusion associated with lethal prostate cancer in a watchful waiting cohort. Oncogene, 2007

Expanding gamut of fusions in prostate cancer [Kumar-Sinha et al. 2009]

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] single gene [Tomlins et al. 2005] none expression and other attributes of individual genes [Kumar-Sinha et al. 2008] networks / modules / gene sets background knowledge employed in discovery

Study: aberrations in the context of interaction networks [Kumar-Sinha et al. 2008]

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] single gene [Tomlins et al. 2005] none expression and other attributes of individual genes [Kumar-Sinha et al. 2008] networks / modules / gene sets background knowledge employed in discovery

Integrating basepair-level and copy number alterations in study of 206 glioblastoma tumors [TCGA 2008]

Recurrent mutually exclusive mutational patterns identified in pathways [TCGA 2008]

Cancer Genome Variation object of discovery pathway / module / gene set [Miller et al. ] [TCGA 2008] [Chin et al. 2006] single gene [Tomlins et al. 2005] none expression and other attributes of individual genes [Kumar-Sinha et al. 2008] networks / modules / gene sets background knowledge employed in discovery

Hallmarks of cancer (“acquired capabilities”) are acquired by positive selection

Hallmarks of cancer (“acquired capabilities”) cause recurrent mutually exclusive mutational patterns

RME (“recurrent mutually exclusive”) pattern discovery algorithm

RME algorithm

Sensitivity and Specificity of the RME algorithm

Applying RME algorithm to glioblastoma

Effect of EP 300 expression on survival in glioblastoma

Laboratory exercise this week: array CGH ( Chia-Chin Wu) Analysis of array CGH data from a set of tumor samples using Genboree – Upload array CGH data – Perform segmentation (invoke Bioconductor tool) – Subtract polymorphisms (databases, current literature) – Identify recurrent amplifications or deletions – Study correlation with gene expression