Approaches to Quantitiative Analysis in the Life Sciences (Graduate)

NJIT MATH 615. Rutgers 48:120:615. 3 Credits.

Now being taught (Spring 2013).

Textbooks

Because of the variety of methods covered, there is no appropriate single textbook, and so one is not required. Students are, however, provided with a list of ‘reference’ texts for various disciplines and applications. Most of these are available in the library, and some are available for consultation from the instructor’s private library. Students are encouraged to select and purchase at least one of the following reference books, when it becomes clear what sort of research they are likely to be doing.

  • Derek A. Roff: An Introduction to Computer Intensive Methods of Data Analysis in Biology. Cambridge UP. Good modern text, covering tools such as maximum likelihood analysis and Bayesian methods as well as traditional hypothesis testing. Code in S-Plus.
  • Alan Grafen and Rosie Hails. Modern Statistics for the Life Sciences. OUP. Another 'modern' approach. Downloadable examples from a variety of packages.
  • Bolker, Benjamin: Ecological Models and Data in R. Princeton University Press. ISBN 0691125228. Ecologicaly focused, but gets you up to speed with R.
  • Kenneth P. Burnham and David R. Anderson. Model Selection and Inference: A Practical Information-Theoretic Approach. Springer. The bible for information-based methods (e.g., AIC and its derivatives), at least in biology.
  • Legendre and Legendre: Numerical methods in Ecology. Mostly advanced stuff like ordination. No code: it's a cookbook for doing things yourself.
  • Ray Hilborn and Marc Mangel: The Ecological Detective. Princeton UP. Partly practical, partly philosophical, it's a nice comaprion to 'cookbook' texts.
  • Manley: Randomization, Bootstrap and Monte-Carlo methods in Biology. Chapman and Hall. Specific to certain techniques, but surprisingly easy to read.
  • Robert Sokal and James Rohlf: Biometry. W. H. Freeman. A classic text, heavily focused on linear regression and ANOVA. Contains 'calculation formulas' and other leftovers form the pre-computer age.
  • Emanuel Paradis: Analysis of phylogenetics and evolution with R (2006). Springer. A narrow focus, but more R examples.
  • Nicholas Gotelli: A Primer of Ecological Statistics. Sinauer. Clearly writen, but not much in the way of code.

Files for download

General Syllabus — Class policies, grading, weekly topics, etc.

Subject Syllabus (PDF) — General listing of topics. This is a rough guide. Timing and order of topics are likely to change. Further down this page is a more detailed syllabus, with dates, that will be updated as the semester progresses.

Concepts and definitions (.doc) — You should download this and begin answering the questions as we cover them in class. Each week I will update it with new questions. You should get the updated version, and copy and paste the new questions to the end of your own version. (Make sure you don't acidentally overwrite the copy that has your answers! I suggest you rename your working copy.)

Template for class project (.doc) — A guide to the structure of your project. I will go through this with you in class.

CDF Player (most recent version) — All students will need this, in order to view and interact with the class files. The only exception is if you already have a working full version of Mathematica. Even in that case, check your version: the files are being revised using v9 of Mathematica, and will work (and look) best in that version. If you can't get v9 of full Mathematica, you should probably get the latest version of the player anyway.

Semester syllabus and weekly downloads

Note: This semester I am converting the Mathematica .nb files to .cdf files, which have lots of added interactivity. The .nb files are from earlier versions of the course. If you want to read ahead, you can use CDF player to view the .nb files! They cover the same material — the only difference is that they won't have as much interactivity. By class time each week a new .cdf will appear that will supercede the old .nb file, and you should grab it.

1/24/2013 — Probability, (lack of) intuition, data types, probability distributions, random numbers
Class 1.cdf

1/31/2013 — Probability distributions continued, quantiles, moments, goodness of fit
Class 2.cdf

2/7/2013 — Fitting models, maximum likelihood, least squares, bias
Class 3.cdf
Matrix multiplication refresher guide

2/14/2013 — Three frameworks for inference: hypothesis testing, model choice, Bayesian
Class 4.cdf

2/21/2013 — ANOVA, linear models, general linear models, generalized linear models
Class 5.nb

Example data for linear regression: resting-R.dat

2/28/2013 — Prediction, cross-validation, overfitting, multiple tests (this contains some material we did not get to in class)
Class 6.nb

Example data for ANOVA: eysenck.txt. Description: Why do older people often seem not to remember things as well as younger people? Do they not pay attention? Do they just not process the material as thoroughly? One theory regarding memory is that verbal material is remembered as a function of the degree to which is was processed when it was initially presented. Eysenck (1974) randomly assigned 50 younger subjects and 50 older (between 55 and 65 years old) to one of five learning groups. The Counting group was asked to read through a list of words and count the number of letters in each word. This involved the lowest level of processing. The Rhyming group was asked to read each word and think of a word that rhymed with it. The Adjective group was asked to give an adjective that could reasonably be used to modify each word in the list. The Imagery group was instructed to form vivid images of each word, and this was assumed to require the deepest level of processing. None of these four groups was told they would later be asked to recall the items. Finally, the Intentional group was asked to memorize the words for later recall. After the subjects had gone through the list of 27 items three times they were asked to write down all the words they could remember. Link for original data: http://www.statsci.org/data/general/eysenck.html
Example data for multiple regression: bodyfat.dat. Description: The data give the body fat, triceps skinfold thickness, thigh circumference and midarm circumference for twenty healthy females aged 20 to 34. The body fat measurement was obtained by an accurate but cumbersome and expensive procedure requiring the immersion of the person in water. It would therefore be very helpful if a regression model with some or all of these predictor variables could provide reliable predictions of the amount of body fat, since the measurements needed for the predictor variables are easy to obtain. Link for original data: http://www.sci.usq.edu.au/staff/dunn/Datasets/applications/health/bodyfat.html

3/7/2013 — Experimental design, types of errors, power
Class 7 Guide.nb

3/14/2011 — Non-parametric statistics
Class 8 Guide.nb

3/21/2013 — SPRING BREAK

3/28/2013 — Ordination, plus recap of Bayesian approach
Class 9 Guide
Bayesian refresher guide.

Class 10 —First R tutorial.

4/4/2013
Class 11 Guide — Autocorrelation and timeseries

4/11/2013 — Second R tutorial.
Example data for second R tutorial: right-click (or CTRL-click on a Macintosh) to download, and save the file somewhere sensible where you can find it again.

4/18/2013 — PRESENTATIONS

4/25/2013 — PRESENTAIONS

5/2/2013 — PRESENTATIONS

5/9/2013 — FINAL EXAM PERIOD. PROJECTS DUE by 6pm!