Introduction to Statistical Data Analysis for the Life Sciences

Any practical introduction to statistics in the life sciences requires a focus on applications and computational statistics combined with a reasonable level of mathematical rigor. It must offer the right combination of data examples, statistical theory, and computing required for analysis today. And it should involve R software, the lingua franca of statistical computing.
Introduction to Statistical Data Analysis for the Life Sciences covers all the usual material but goes further than other texts to emphasize:


Both data analysis and the mathematics underlying classical statistical analysis
Modeling aspects of statistical analysis with added focus on biological interpretations
Applications of statistical software in analyzing real-world problems and data sets


Developed from their courses at the University of Copenhagen, the authors imbue readers with the ability to model and analyze data early in the text and then gradually fill in the blanks with needed probability and statistics theory. While the main text can be used with any statistical software, the authors encourage a reliance on R. They provide a short tutorial for those new to the software and include R commands and output at the end of each chapter. Data sets used in the book are available on a supporting website.

Each chapter contains a number of exercises, half of which can be done by hand. The text also contains ten case exercises where readers are encouraged to apply their knowledge to larger data sets and learn more about approaches specific to the life sciences. Ultimately, readers come away with a computational toolbox that enables them to perform actual analysis for real data sets as well as the confidence and skills to undertake more sophisticated analyses as their careers progress.

Claus Thorn Ekstrøm is an associate professor of statistics in the Department of Basic Sciences and Environment and leader of the Center for Applied Bioinformatics in the Faculty of Life Sciences at the University of Copenhagen. His research interests include genetic marker error detection, simulation-based inference, image analysis, and the analysis of microarray DNA chips, metabolic profiles, and quantitative traits for complex human families.
Helle Sørensen is an associate professor of statistics and probability theory in the Department of Mathematical Sciences in the Faculty of Science at the University of Copenhagen. Her research interests include statistical applications in eco-toxicology and animal science as well as statistical methods for stochastic processes.

Description of Samples and PopulationsData typesVisualizing categorical dataVisualizing quantitative dataStatistical summariesWhat is a probability?Linear RegressionFitting a regression lineWhen is linear regression appropriate?The correlation coefficientPerspective
Comparison of GroupsGraphical and simple numerical comparisonBetween-group variation and within-group variationPopulations, samples, and expected valuesLeast squares estimation and residualsPaired and unpaired samplesPerspectiveThe Normal DistributionPropertiesOne sampleAre the data (approximately) normally distributed?The central limit theorem
Statistical Models, Estimation, and Confidence IntervalsStatistical modelsEstimationConfidence intervalsUnpaired samples with different standard deviationsHypothesis TestsNull hypothesest-testsTests in a one-way ANOVAHypothesis tests as comparison of nested modelsType I and type II errorsModel Validation and PredictionModel validationPredictionLinear Normal Models Multiple linear regressionAdditive two-way analysis of varianceLinear modelsInteractions between variablesProbabilitiesOutcomes, events, and probabilitiesConditional probabilitiesIndependenceThe Binomial DistributionThe independent trials modelThe binomial distributionEstimation, confidence intervals, and hypothesis testsDifferences between proportionsAnalysis of Count DataThe chi-square test for goodness-of-fit2 × 2 contingency tableTwo-sided contingency tablesLogistic Regression Odds and odds ratiosLogistic regression modelsEstimation and confidence intervalsHypothesis testsModel validation and predictionCase Exercises Case 1: Linear modelingCase 2: Data transformationsCase 3: Two sample comparisonsCase 4: Linear regression with and without interceptCase 5: Analysis of variance and test for linear trendCase 6: Regression modeling and transformationsCase 7: Linear modelsCase 8: Binary variablesCase 9: AgreementCase 10: Logistic regression
Appendix A: Summary of Inference Methods Statistical conceptsStatistical analysisModel selection
Appendix B: Introduction to R Working with RData frames and reading data into RManipulating dataGraphics with RReproducible researchInstalling RExercises
Appendix C: Statistical TablesThe x2 distributionThe normal distributionThe t distributionThe F distribution

Bibliography
Index
R Commands and Output and Exercises appear at the end of each chapter.