We think you have liked this presentation. If you wish to download it, please recommend it to your friends in any social system. Share buttons are a little bit lower. Thank you!
Presentation is loading. Please wait.
Published byDallin Gibbons
Modified over 2 years ago
12 Jan 2006Hort Retreat © Brian S Yandell1 Genetics, Microarrays & Evolution: Issues as a Statistician (sans formula) Brian S. Yandell Horticulture, Statistics & Biometry University of Wisconsin-Madison
12 Jan 2006Hort Retreat © Brian S Yandell2 who am I (professionally)? Professor Brian S. Yandell joint appointment across colleges: –50% Horticulture (CALS) –50% Statistics (Letters & Sciences) –UW-Madison since 1982 Biometry Program teaching & research
12 Jan 2006Hort Retreat © Brian S Yandell3 biometry program MS degree coadvise with biologist bridge biology & stats project & oral report consulting experience 10 completed, 1 current Botany, Dairy Sci (2), Genetics, Hort, Land Resources, Meat & Animal Sci, Wildlife Ecology (2), Zoology consulting facility statistical consulting –5 faculty, 2-4 students computing assistance –2 staff + operators self-help model –guide research ideas –build skill sets collaboration –students, faculty, staff –CALS, VetMed, L&S
12 Jan 2006Hort Retreat © Brian S Yandell4 research & teaching statistical genetics –QTLs in Brassica & mice –genetic architecture gene action epistasis –microarrays differential expression genetical genomics statistical ecology –population ethology –individual-based simulations stats consulting –communication skills –write, plot, talk –bridge stats & biology linear models –experimental design –complicated analysis –problems directly from consulting –published textbook
12 Jan 2006Hort Retreat © Brian S Yandell5 what is statistics? We may at once admit that any inference from the particular to the general must be attended with some degree of uncertainty, but this is not the same as to admit that such inference cannot be absolutely rigorous, for the nature and degree of the uncertainty may itself be capable of rigorous expression. — Sir Ronald A. Fisher (1935 The Design of Experiments) digital.library.adelaide.edu.au/coll/special/fisher
12 Jan 2006Hort Retreat © Brian S Yandell6 what is biology? Biology … consists of two rather different fields, mechanistic (functional) biology and historical [evolutionary] biology. Functional biology deals with … cellular processes, including those of the genome. … [Evolutionary biology] involve[s] the dimension of historical time. — Ernst Mayr at 100 (What Makes Biology Unique? 2004 Cambridge U Press)
12 Jan 2006Hort Retreat © Brian S Yandell7 Doonesbury by Garry Trudeau 18 dec 2005 www.doonesbury.com
12 Jan 2006Hort Retreat © Brian S Yandell8 functional biology: how does life work? broad scientific questions –how do plants modify flowering time? –how do mice (humans) develop diabetes? molecular investigations –genetic association: QTL(s) –fine mapping: candidate gene(s) –biochemical pathways: causal models
12 Jan 2006Hort Retreat © Brian S Yandell9 DNA RNA protein metab- olites J Watson & F Crick (1953) R Franklin (1953) www.accessexcellence.org/RC/VL/GG/central.html
12 Jan 2006Hort Retreat © Brian S Yandell10 central dogma via microarrays (Bochner 2003)
12 Jan 2006Hort Retreat © Brian S Yandell11
12 Jan 2006Hort Retreat © Brian S Yandell12 genetical genomics: mapping microarrays (Jansen Nap 2001)
12 Jan 2006Hort Retreat © Brian S Yandell13 intercross: genetic mosaic of P1,P2 (from KW Broman) QTL
12 Jan 2006Hort Retreat © Brian S Yandell14 coordinated expression in mouse genome (Schadt et al. 2003) expression pleiotropy in yeast genome (Brem et al. 2002)
12 Jan 2006Hort Retreat © Brian S Yandell15 SCD1: epistatic interaction plots (chr 5x9 p=0.007; chr 2x9 p<0.0001) mostly axdmostly axa
12 Jan 2006Hort Retreat © Brian S Yandell16 genetic architecture for mRNA (SCD1) expression Model: Df MS F Pr(F) Model 7 69.060 73.10 0.00000 *** Error 99 0.945 Total 106 70.005 Single term deletions Df MS F Pr(F) Chr2@80 1 9.073 9.60 0.0025 ** Chr2@105 1 0.073 0.08 0.78 Chr5@67 1 0.218 0.23 0.63 Chr9@67 1 7.156 7.57 0.0070 ** Chr9@67dom 1 0.106 0.11 0.74 Chr2@105:Chr9@67 1 15.612 16.52 0.000096 *** Chr5@67:Chr9@67dom 1 7.211 7.63 0.0068 **
12 Jan 2006Hort Retreat © Brian S Yandell17 SCD1: marginal* LOD by locus green = 1-QTL scan (add + dom) blue = additive red = dominance purple = epistasis black = total epistasis! *: effect of 1 QTL adjusting for all other QTL
12 Jan 2006Hort Retreat © Brian S Yandell18 SCD1: marginal 2-QTL scan examine 2 QTL in presence of others 2-QTL only scanmarginal 2-QTL scan
12 Jan 2006Hort Retreat © Brian S Yandell19 how to assess genetic architecture via genetical genomics? screen mRNA across segregating panel –which show strong evidence of heritability? organize into functional groups –correlation across panel –(Lan, Chen et al. 2006 PLoS Genetics) infer genetic architecture by group –allow for multiple QTL and epistasis validate –comparative genomics, KOs, etc. infer biochemical pathways
12 Jan 2006Hort Retreat © Brian S Yandell20 trait: log10(insulin) at 10 weeks in mice QTL on Chr 19 solid=512 mice (now) dashed=311 mice (then) black=all blue=male red=female purple=sex-adjusted from QTL to candidate gene?
12 Jan 2006Hort Retreat © Brian S Yandell21 candidate gene: Sorcs1 meta-analysis across 11 sub-congenic strains marker regression & within-strain permutations Clee et al. (2006 in review) fine mapping with meta-analysis
12 Jan 2006Hort Retreat © Brian S Yandell22 Sorcs1 gene & SNPs (only gene in region with SNP variation between parent strains)
12 Jan 2006Hort Retreat © Brian S Yandell23 Sorcs1 study in humans 5 haplotype blocks
12 Jan 2006Hort Retreat © Brian S Yandell24 candidate genes in QTL regions lab experiments on pathway components graphical models via genetical genomics goal: unravel biochemical pathways (with Elias Chaibub) R2 D2 P2 QTL R1 D1 P1 observable trans-action observable cis-action?
12 Jan 2006Hort Retreat © Brian S Yandell25 evolutionary biology: why did life end up this way? natural selection –why do certain species persist/perish? –why do some characteristics emerge/persist/perish? modeling individuals in a population –historical narratives –time-based simulations –event-based simulations
12 Jan 2006Hort Retreat © Brian S Yandell26 what is natural selection? Darwin’s “variation and selection” 1. random variation mutation, meiosis, gamete meeting 2. differential selection selection of the best (fittest) culling/elimination of the worst selection on individuals, not populations survival selection sexual selection (reproductive success) — Ernst Mayr at 100 (What Makes Biology Unique? 2004 Cambridge U Press)
12 Jan 2006Hort Retreat © Brian S Yandell27 individual-based models in population ethology www.ipm.ucdavis.edu/PHENOLOGY/models.html California red scale damage to orange two parasitoids dead red scale
12 Jan 2006Hort Retreat © Brian S Yandell28 event-driven model: red scale life history
12 Jan 2006Hort Retreat © Brian S Yandell29 what do we measure? trait: aspect of an individual that is constant or slowly/rarely changing over span of study event: significant biological change in state that is instantaneous at resolution of study types of events –birth, death –change of “health” or “development” state flower opening, seed setting, dormancy –interactions with other individuals predation, parasitism, plant harvesting reproduction, pollination, fruit/seed harvesting (Bland Ewing et al. 2002 Ecol Model)
12 Jan 2006Hort Retreat © Brian S Yandell30 how to incorporate genetics? genotype is a trait of an individual differential probabilities for events –depending on genotype, event history –change event rate or shape 0 1 second instar third instar
12 Jan 2006Hort Retreat © Brian S Yandell31 what is our scope of measurement? measurements are in context of span & resolution –mechanics of measuring –focus of key questions in time and space span: largest amount of time/space studied, with aspects over longer intervals considered constant or slowly varying resolution: smallest increment of time/space contributing useful biological information, with processes over smaller scales assumed to be instantaneous (Bland Ewing et al. 2002 Ecological Modeling) resolution (1 day) span (1 msec) orange: span (1 year) resolution (1 picosec)molecular:
12 Jan 2006Hort Retreat © Brian S Yandell32 modeling natural selection depends on genetics, event history, environment differential survival –chance of surviving parasitism –development rate –ability to migrate to new habitat sexual selection –mate preference based on genotype –other gene flow vectors (co-evolution) –barriers to reproduction (speciation)
12 Jan 2006Hort Retreat © Brian S Yandell33 how do we infer general properties from historical narrative simulations? cannot predict exact outcomes –probability over range of scenarios –variability across multiple simulations “black box” testing –sensitivity to changes in conditions temperature, initial population sizes rates of occurrence of life events –comparison with historical records
12 Jan 2006Hort Retreat © Brian S Yandell34 summary who am I? what do statisticians do? what is biology? how have I studied genetics? how can microarrays deepen our models? how might I study evolution?
Professor Brian S. Yandell joint faculty appointment across colleges: –50% Horticulture (CALS) –50% Statistics (Letters & Sciences) Biometry Program –MS.
13 October 2004Statistics: Yandell © Inferring Genetic Architecture of Complex Biological Processes Brian S. Yandell 12, Christina Kendziorski 13,
Yandell © June Bayesian analysis of microarray traits Arabidopsis Microarray Workshop Brian S. Yandell University of Wisconsin-Madison
Yandell © 2003Wisconsin Symposium III1 QTLs & Microarrays Overview Brian S. Yandell University of Wisconsin-Madison
Yandell © 2003JSM Dimension Reduction for Mapping mRNA Abundance as Quantitative Traits Brian S. Yandell University of Wisconsin-Madison
Quantile-based Permutation Thresholds for QTL Hotspots Brian S Yandell and Elias Chaibub Neto 17 March © YandellMSRC5.
The genetic dissection of complex traits
Chapter 22 - Quantitative genetics: Traits with a continuous distribution of phenotypes are called continuous traits (e.g., height, weight, growth rate,
Evolution of Populations
AP Biology Heredity PowerPoint presentation text copied directly from NJCTL with corrections made as needed. Graphics may have been substituted with a.
Unit 7: Evolution.
What is a QTL? Quantitative trait locus (loci) Region of chromosome that contributes to variation in a quantitative trait Generally used to study “complex.
Natural Selection Developed by Charles Darwin in 1859
Methods of Genome Mapping linkage maps, physical maps, QTL analysis The focus of the course should be on analytical (bioinformatic) tools for genome mapping,
Essential Question: How can a change in the environment initiate a change in the population? NATURAL SELECTION.
Evolution Species Changing over time. Charles Darwin Evolution by Means of Natural Selection.
Bio 7: General Biology II Evolutionary, Organismal, & Ecological Biology Dr. Diane Livio myetudes.org/portal MW 2:30-4:30 (CMS 229)
Evolution and Gene Frequencies
© 2017 SlidePlayer.com Inc. All rights reserved.