Transcript Slide 1
Exploring Cancer Incidence Rates: the Multi-hit Model of Cancer in STELLA Kam Dahlquist Biology Loyola Marymount University Seamus Lagan Physics Jeff Lutgen Math Whittier College BioQUEST: Investigating Interdisciplinary Interactions June 18, 2005 Audience for this Exercise Biology: Non-majors Introductory Biology (BIOL 201: Cell Function) Genetics, Cell Biology, Molecular Biology Bioinformatics Math: Mathematical Modeling Differential Equations Calculus Probability & Statistics Cancer as a Theme in BIOL 201: Cell Function A new literary metaphor for the genome: Dramatis Personae --proteins of the cell cycle recast as Romeo & Juliet --“superpowers” of cancer Write a “Perspectives” article and give a poster presentation about a primary research article about a “cancer gene” (pre-genomics era) MAPPFinder analysis of DNA microarray data from prostate cancer (genomics era) Modeling of Cancer Incidence Rates (this exercise) Incidence of Colon Cancer in Different Age Groups What is the shape of this plot? What does it mean? Why is the shape like this? http://www.cancerquest.org/index.cfm?page=302 Can we create a model in STELLA that will reproduce the main features of this plot? The members of our team contributed: Kam: The biology of the system and reasonable values to use for the parameters Seamus: the model in STELLA Jeff: a Java program that will run the same model as STELLA thousands of times to collect a large dataset and display results The Biology of Cancer The multi-hit model: a cell needs to accumulate 4 – 7 independent mutations in “cancer causing” genes to become cancerous Proto-oncogenes: genes whose normal function is to stimulate the cell cycle and/or prevent cell death; only one allele needs to be mutated to lead to cancer Tumor suppressors: genes whose normal function is to inhibit the cell cycle and/or stimulate cell death; both alleles need to be mutated to lead to cancer What do we need to know? Where can we find the information? Reasonable Inputs to the Model Average length of a gene in the human genome (open reading frame, excluding introns) frequency Average protein = 457 amino acids 457 X 3 = ~1300 nucleotides protein length http://www.ebi.ac.uk/integr8/StatsLengthPage.do?orgProteomeID=25 Reasonable Inputs to the Model Rate of mutation: 1 nucleotide in 1 billion per cell division estimate from Freeman’s Biological Sciences text Number of proto-oncogenes: 279 Number of tumor suppressors: 67 according to the Cancer Gene Census list http://www.sanger.ac.uk/genetics/CGP/Census/ Follow one cell as it divides at a rate of one cell division per day Equilibrium model: for each cell division, one cell dies Three Models were Built in STELLA Took into account: --Role of tumor suppressors and proto-oncogenes --The number of mutations accumulated before the onset of cancer Model 1: no tumor suppressors in model, no tumor suppression or DNA repair, when 4 different proto-oncogenes are mutated, then cancer results Model 1, Run 1 Model 1, Run 2 Three Models were Built in STELLA Model 2: tumor suppressor genes will suppress the formation of tumors no matter how many proto-oncogenes are mutated. This continues until both alleles of at least one of the tumor suppressor genes are mutated, at which time all suppression ceases and tumors are free to form if there are enough mutated proto-oncogenes. Model 2, Run 1 Model 2, Run 2