# Register now to access 7 million high quality study materials (What's Course Hero?) Course Hero is the premier provider of high quality online educational resources. With millions of study documents, online tutors, digital flashcards and free courseware, Course Hero is helping students learn more efficiently and effectively. Whether you're interested in exploring new subjects or mastering key topics for your next exam, Course Hero has the tools you need to achieve your goals.

16 Pages

### lecture_5

Course: BI 800, Fall 2009
School: Michigan
Rating:

Word Count: 1085

#### Document Preview

800 Bioinformatics Module 2 Lecture 5 March 17, 2009 Kerby Shedden Department of Statistics kshedden@umich.edu 1 Review of simple hypothesis testing Identify a null-hypothesis (that nothing interesting is going on) and (optionally) a set of alternative hypotheses of interest. Construct a test statistic T from the data. T should have these properties: T is close to zero when the data support the null...

Register Now

#### Unformatted Document Excerpt

Coursehero >> Michigan >> Michigan >> BI 800

Course Hero has millions of student submitted documents similar to the one
below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.

Course Hero has millions of student submitted documents similar to the one below including study guides, practice problems, reference materials, practice exams, textbook help and tutor support.
800 Bioinformatics Module 2 Lecture 5 March 17, 2009 Kerby Shedden Department of Statistics kshedden@umich.edu 1 Review of simple hypothesis testing Identify a null-hypothesis (that nothing interesting is going on) and (optionally) a set of alternative hypotheses of interest. Construct a test statistic T from the data. T should have these properties: T is close to zero when the data support the null hypothesis; T gets further from zero when the data support the alternative hypothesis. Under the null hypothesis, if we repeatedly sample independent data sets and recalculate T for each set, the variation in T should follow a known distribution (usually standard normal). Suppose Tobs is the observed value of T . The p-value is the probability that |T | is greater than |Tobs | if the null hypothesis is true. Small p-values imply more evidence in the data against the null hypothesis, larger p-values imply more evidence in the data in support of the null hypothesis. 2 Errors in hypothesis testing A "false positive" occurs when the null hypothesis is true, but is rejected. This is also called a "type-I" error. A "false negative" occurs when the alternative hypothesis is true, but the null hypothesis is not rejected. This is also called a "type-II" error. Classical hypothesis testing is set up so that false positives occur a fixed fraction of the time (usually 5%). The probabilty of a false negative occuring depends on the effect size, and is equal to 1 minus the power. 3 Multiple testing Suppose we test a family of hypotheses, e.g. we collect 10 possible biomarkers for lung cancer recurrence, and we perform these tests: Marker 1 does (alternative) or does not (null) associate with recurrence. Marker 2 does (alternative) or does not (null) associate with recurrence. Marker 3 does (alternative) or does not (null) associate with recurrence. If perform all 10 of these tests, and if the null hypotheses are all true, and if the test statistics are independent, the probability that we make at least one false positive statement is around 0.4. This is called the "family-wise error rate." false-positive rate (which is usually 5%). It is always greater than the 4 Examples of multiple testing Genome-wide association studies (GWAS) test up to several million genetic variants for association with a trait. Molecular library screening test a large set of compounds for activity in a high throughput assay (for enzyme inhibition, toxicity, binding to a receptor, ...). Differential gene expression analysis test a large number of genes for differential expression between two classes of samples (e.g. disease vs. healthy). Combination therapy evaluation test various dose combinations of several drugs for safety or efficacy. Biomarker evaluation test a panel of biomarkers for diagnostic or prognostic performance. 5 Multiple testing When many tests are performed, we can ask the following questions: Are there any true positives? How many true positives are there? Which are the true positives? How you approach things should depend on your goals: surveillance, discovery, validation, estimation, . . . 6 Scales of multiple testing "Classical multiple testing": typically, < 10 tests are performed looking at different treatment contrasts, dose combinations, etc. "Large scale multiple testing": testing all genes, all proteins, etc. 7 Answering the question "are there any true positives?" For large scale testing, the simplest approach is to assess the uniformity of the p-values. If all the null hypotheses are true, test the statistics should be uniformly distributed in the interval (0,1). 8 One way of seeing that the p-values are uniformly distributed when all tested items are truly negative is to plot the "empirical CDF" of the p-values: 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.20 m=100 20 Expected value Expected value 2SD 40 Rank 60 80 100 1.2 1.0 0.8 0.6 0.4 0.2 0.0 -0.20 m=1000 p p 200 Expected value Expected value 2SD 400 Rank 600 800 1000 9 Quantile analysis of the test statistics 1000 independent tests, 20 non-nulls (effect size = ) on left, zero non-nulls on right. Ten replications plotted in each graph. 6 Quantiles of test statistic 2 0 -2 -4 -6-6 Standard normal quantiles -4 -2 0 2 4 Quantiles of test statistic 4 6 4 3 2 1 0 -1 -2 -3 -4-4 -3 Standard normal quantiles 10 -2 -1 0 1 2 3 4 Controlling family-wise error: Bonferroni approach Without assuming independence of the tests, we get the inequality P0 (T1 > T or T2 > T or Tm > T ) m. So if we carry out each test at level = /m then the family-wise error rate is at most . If = 0.05 and m = 100 then = 0.0005. Simple recipe: multiply your p-values by the number of tests (and round down to 1 if the result is greater than 1). 11 False Discovery Rate Suppose we have test statistics Z1 , Z2 , . . . , Zm (i.e. one test per gene in a GWAS, or one test per compound in a screen for enyzme inhibitors). We can sort the test statistics and specify a "tail region" that defines our objects (genes, compounds) of interest. Within the tail region, there is a certain number of "true positive" and "false positive" calls. These are denoted as TP:FP below. -8 -8 -8 -6 -6 -6 -4 -4 -4 -2 -2 -2 0 0 0 10000:45500, FDR=0.82 6 8 2 4 5000:2700, FDR=0.28 6 8 2 4 1000:63, FDR=0.06 6 8 2 4 The false discovery rate is F P/(T P + F P ) where F P is the number of false positive calls and T P is the number of true positive calls. 12 Estimating the False Discovery Rate The FDR is defined in terms of FP and TP, but we only know FP+TP from our data. To estimate the FDR we have to use the statistical properties of our test (i.e. that the test ...

Find millions of documents on Course Hero - Study Guides, Lecture Notes, Reference Materials, Practice Exams and more. Course Hero has millions of course specific materials providing students with the best way to expand their education.

Below is a small sample set of documents:

Michigan - BI - 800
Bioinformatics 800 Module 2 Lecture 4March 11, 2008Kerby Shedden Department of Statistics kshedden@umich.edu1Pearson (product-moment) correlation Start with paired measurements X, Y : e.g. width,height; education,income; -catenin protein expres
UPenn - MEAM - 550
MEAM 550 Problem 1z i +1 =Fall 2001 Solution #5University of Pennsylvania AV 2 2k ( g 0 z i ) 2Begin with i = 0, z 0 = 0 and proceed until difference between two successive zs is less than a tolerance value. This is the simple principle of
UPenn - MEAM - 550
C. T.-C. Nguyen, Micromechanical components for miniaturized low-power communications (invited plenary), Proceedings, 1999 IEEE MTT-S International Microwave Symposium RF MEMS Workshop (on Microelectromechanical Devices for RF Systems: Their Construc
UPenn - MEAM - 550
UPenn - MEAM - 550
MEAM 550Modeling and Design of MEMSSpring 2004Solution to homework #6 Problem 1p2L VFlexureL p3LNarrow beam ConnectorT0 p2L p1Lp2Lp2L(1-p1)L KsWide beamOut-of-plane thickness= ptLFig. 1 Basic electro-thermal microactuato
UPenn - MEAM - 550
MEAM 550 Problem 1 KVL for the circuit gives m + kz + bz1 = z k a ( z - z1 ) = bz12 0 AVin 2( g 0 - z ) 2Fall 2001 Solution #7University of Pennsylvania By defining the state variables x1 = z , x 2 = z , x3 = z1 , the state equations can be
UPenn - MEAM - 550
Mechanical Engineering and Applied MechanicsUniversity of PennsylvaniaA glimpse of MEMSPresented to MEAM 550 (Fall 2001) studentsG. K. AnanthasureshSeptember 17, 2001G.K. Ananthasuresh, U. of Pennsylvania, Sep. 2001What's in a name?Mic
UPenn - MEAM - 550
HOMEWORK #1 FOR MEAM550ZHONGYAN QIAN 1. Answer: Aluminum's material properties are: Density: = 2.7 10 3 kgGPam3Modulus: E = 69Material thermal conductivity: K = 222 W Material resistivity: = 2.83 10 -8 m Water's material properties are:
UPenn - TCOM - 501
TCOM501 Networking: Theory &amp; Fundamentals Midterm Examination Professor Yannis A. Korilis March 6, 2002Answer all problems. Good Luck! Problem 1 [15 points]: 1. State the PASTA theorem. [5 points] 2. Provide an analytical proof of the PASTA theore
UPenn - STAT - 955
Stochastic Calculus and Financial Applications Final Take Home Exam (Fall 2006) SOLUTIONSInstructions. You may consult any books or articles that you find useful. If you use a result that is not from our text, attach a copy of the relevant pages fro
Lecture Notes for AI (CS-482) Fall 2008, Lecture 23Steven Christman November 12, 20081Partial Observable Markov Decision Processes1. Exact method - which is to turn the problem into an MDP. This is done by converting the states to belief states
Computer Science and Engineering Department University of Nevada, RenoCS 482/682 Artificial IntelligenceLecture 25 Ensemble Learning &amp; Maximum Likelihood LearningIn which we view learning as a form of uncertain reasoning from observationsEnsemb
UPenn - ENM - 511
Haim H. Bau (edited by J. L. Bassani 2/28/01)1/20Numerical Solutions to ODEs and PDEs via Maple and Matlab - A Heat Transfer Application EXTENDED SURFACES: FIN ANALYSIS - A CASE STUDY In many heat transfer applications, it is desirable to incre
Penn State - BUB - 1945
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/45High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::75 Rank (1=Warm
Penn State - BUB - 1937
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/37High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth39 Rank (1=Warmest, 42=
Penn State - BUB - 1952
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/52High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth40 Rank (1=Warmest, 57=
Penn State - BUB - 1989
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/89High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::50 Rank (1=Warmest,
Penn State - BUB - 1954
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/54High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::46 Rank (1=Warm
Penn State - BUB - 1957
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/57High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::55 Rank (1=Warm
Penn State - BUB - 1968
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/68High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::47 Rank (1=Warmest,
Penn State - BUB - 1972
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/72High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::47 Rank (1=Warmest,
Penn State - BUB - 1960
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/60High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::31 Rank (1=Warmest,
Penn State - BUB - 1946
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/46High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::53 Rank (1=Warm
Penn State - BUB - 1941
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/41High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::33 Rank (1=Warm
Penn State - BUB - 1959
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/59High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::41 Rank (1=Warmest,
Penn State - BUB - 1942
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/42High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::68 Rank (1=Warm
Penn State - BUB - 1944
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/44High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::53 Rank (1=Warm
Penn State - BUB - 1971
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/71High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::47 Rank (1=Warmest,
Penn State - BUB - 1934
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/34High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth67 Rank (1=Warmest, 39=
Penn State - BUB - 1931
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/31High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth37 Rank (1=Warmest, 36=
Penn State - BUB - 1948
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [Midnight - Midnight LT] -03/17/48High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::49 Rank (1=Warm
Penn State - BUB - 1961
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/61High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::31 Rank (1=Warmest,
Penn State - BUB - 1962
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/62High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::36 Rank (1=Warmest,
Penn State - BUB - 1973
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/73High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth: ::63 Rank (1=Warmest,
Penn State - BUB - 1982
Sheet1 STATE COLLEGE, PENNSYLVANIA DAILY WEATHER SUMMARY - Latest Complete Day [1200 UT - 1200 UT] -03/17/82High Temperature Low Temperature Mean Temperature Rain or Liquid Equivalent Snow and/or Ice Pellets Snow Depth40 Rank (1=Warmest, 87=Colde
Penn State - GCB - 5013
Project Design A.J. Zambanini Greg BeniniPurposeTo create a one man semi-automated shooting trap that is: Efficient Effective Safe Affordable Simply Designed Aim: To fill the gap between manual traps and expensive fully automatic traps.What it D
Penn State - ECM - 5028
EL I ZAB E T H C. MELESHENKOSC H O O L AD D R E S S : 418 H I B BS H A L L , U N I V E RS I T Y PARK, PA 16802 H O M E A D D R E S S : 201 AV E N U E A, PIT TSB U RG H , PA 15221 P H O N E ( 4 1 2 ) 3 7 7 - 8 7 7 1 E - M A I L E C M
UPenn - EC - 714
How to Access Standard Data SourcesClaudia Olivetti First Version, September 19 1998 This Version September 15, 19991IntroductionThese notes describe a few sources of data that are easily available to Econ Graduate students in Penn. The notes
UPenn - EC - 714
(?), (?), (?). (?), (?), (?), (?). (?) (?). (?). (?). (?) (?) (?) (?) (?) (?) (?) (?) (?) (?) (?) (?). (?) (?) (?) (?) (?) (?) (?). (?) (?) (?) (?) (?) (?) (?) (?). (?). (?). References Aiyagari, S. R., Greenwood, J., and Gner, N. 1997. The state of
Penn State - ASPRING - 471
Problem 47. In a ABC inscribed a rhombus ADEF such that the angle A is joint, D is on AB, E is on BC, and F is on AC. Find the side of the rhombus if |AB|=c and |AC|=b. Problem 46. Let ABCD be a trapezoid AD BC. Let |AD|=20,|BC|=10, |AC|=18. Let O be
Michigan - NRE - 701
MASTERS PROJECTS PLANNING COURSE NRE 701-888WINTER 2003 ASSIGNMENT #1CRITIQUE OF PAST MASTERS PROJECT REPORTSAssigned January 6, Due January 27Developing a Master's Project is easier when you have an idea of the forms a project can take. Your as
Maggie Creek DistrictGEOLOGY AND MINERALIZATION OF THE MAGGIE CREEK DISTRICTJ.B. Harlan1, D.A. Harris1, P.M. Mallette1, J.W. Norby2, J.C. Rota1, and J.J. Sagar1ABSTRACTFirst organized in the late 1870s, the Maggie Creek district is one of the l
UPenn - MATH - 360
Series Series of Nonnegative Terms The Number eSeriesAll sequences/series will be complex valuedDefinitionGiven a sequence {an } we associate with the sequence {an } a sequence {sn } wherensn =i=1ai = a1 + a2 + a3 + + anWe call this
UPenn - CIS - 501
Lecture 9: Memory Hierarchy (3)Michael B. Greenwald Computer Architecture CIS 501 Fall 1999MBG 1CIS501, Fall 99Improving Cache Performance Average memory-access time = Hit time + Miss rate x Miss penalty (ns or clocks) Improve performance by:
UPenn - HIST - 349
HIST 349-401 History of Sexuality in the United StatesCourse information: Fall 2007 Tues/Thurs. 1:30-3:00 Room: Arch/Crest Contact information: Professor Kathy Peiss, peiss@sas.upenn.edu Office: 315C College Hall; phone 215-898-2746 Office hours: We
The People's Republic of China Under Chairman Mao ZedongHistorical Background: Longest continuous civilization: at least 4000 years of good history. Empire unified in 221 BC from warring states Dynastic Cycle: stable imperial autocracy divided by pe
Laser Tweezers and Laser ScissorsBrian Hull Physics 461 March 21st, 2004Laser Tweezers and Scissors BeginningsLaser scissors came firstDonald E. Rounds and Michael W. Berns wrote a paper in Scientific American, february 1970 titled Cell Surgery
UPenn - EXEN - 695
EMTM 695V. KumarAssignment 41. Read the Miniature Robos for Space and Military Missions article in Section 10 of the Bulkpack. Identify three common technological needs of space and military robots.a) Autonomy in navigation: Planetary robots
UPenn - CIS - 650
Reconciling Schemas of Disparate Data Sources: A Machine-Learning ApproachAnHai Doan, Pedro Domingos, Alon Halevy
People's Republic of China After Chairman MaoqqqqIn 1972, Mao Zedong met with Nixon and relations with the rest of the world began to normalize. Mao Zedong, Zhu De, and Zhou Enlai died in 1976. Hua Guofeng was new Chairman, and arrested &quot;Gan
Penn State - ME - 340
PENNSTATEH. J. Sommer III, Ph.D., Professor Department of Mechanical and Nuclear Engineering The Pennsylvania State University 337 Leonhard Building University Park, PA 16802 Office: 814-863-8997 Fax: 814-865-9693 hjs1@psu.edu http:/www.mne.psu.edu/
UPenn - STAT - 712
Statistics 712 Spring, 1999Lecture 26 1Introduction to Game TheoryAdministrative Things Assignment 8 due Friday. Questions? Another reference on game theory is W. Poundstone, Prisoner's Dilemma (1992). Poundstone's book includes some of the his
UPenn - CSE - 240
CSE 240: Introduction to Computer Architecture09/06/2005 10:38 AMCSE 240: Introduction to Computer Architecture (Autumn 2005)Mon/Wed/Fri 12:00 - 12:50 Towne Heilmeier Hall (CLASS SCHEDULE) Email: cse240-001-05c@lists.upenn.edu (archive) Web: htt
Penn State - KMK - 398