BIO/CSC295 2011F, Class 01: Getting Started Admin: * Introductory Survey assigned, due Tuesday. * Please fill out the RISC Survey asap. (Link on class page. Sam will also email.) http://www.grinnell.edu/academic/psychology/faculty/dl/risc/ * Three readings from Science distributed (links on Pioneerweb). Read and respond for Tuesday's class. See the notes on response papers for information on writing responses * EC for attending the 39 Steps in Central Park * Tonight at 7:30 * Saturday at 7:30 * Sunday at some time * EC for New Student Disorientation Concert Saturday at 10ish in Gardner * EC for Football Scrimmage Saturday at 1pm Overview: * Survey. * Introductions. * What is Bioinformatics? * About the Course. * Discussion of Wired Dinosaur DNA Reading. * Getting Started with the Linux Network. * Getting Started with Python. The joy of getting passwords for MathLAN Introductions * A few math majors * Lots of bio majors * A few undeclared people * A variety of perspectives, from ecology to neuroscience to "I wish I wasn't a biology major" to computational chemistry to English to music to dabbling to deep understanding philosophical underpinings * Some people have done a bit of bioinformatics * Some people mistakenly think this will fun * People know that both Biology and CS are cool and want more * People recognize the value of approaching things from multiple perspectives What is Bioinformatics? * Using computational methods to solve biological problems. * What are computational methods? Computers, math, and statistics. * What are biological problems? (How do we get the goats away from the valuable plants?) * Or perhaps a subset * Looking for patterns * Dealing with huge amounts of data (e.g., human genome, protein fragments) * Model complex systems * Play with data * Epidemiology * What tools have you used when you did bioinformatics? * Vector-thing matches gene sequences * Extrapolate hard data into unknown areas, which you can then explore experimentally * Phylogenetic trees * And application to studying human populations * BLAST - Basic Local Alignment Search Tool Put a DNA sequence into your computer and click a button that says search, and the ghost in the machine says "These known sequences are pretty damn close to what you have." * Are BLAST searches bioinformatics? * Some say yes - It's using computational tools to look at biological data * Some say no - Using tools is not the same as building tools * Some refuse to answer * Congratulations, you are replicating a typical debate * If using tools is not bioinformatics, then what is bioinformatics? * Being able to write your own tools or to modify existing tools * Example: Mapping lots of short segments that are then mapped to the genome. But someone else has to do the statistical analysis. * Not-so-hidden moral in Professor Praitis's statement: Good bioinformatics is often COLLABORATIVE: People work together, drawing upon their own skill sets, to build something that they could not do individually. * And that's the model that we're going to follow in the class. * To collaborate, you need to know something of the other person's language * Biologists need some experience with seeing how computer scientists approach the world * Computer scientists need to understand the underlying biologic concepts * Our model for the class * Use Web tools * Write your own programs ("get messy") * Wet labs ("get really messy") * And do it all together so that you can communicate across disciplines! * Wired Article * Summary * Some scientists claimed that they extracted proteins from a dinosaur bone AND that the proteins bore remarkable similarity to proteins in that chickens have * There is a controversy about this * Discussion * How do we know that the claim is true? * It MUST be true. It's published in Science. * Why might it not be true? * The data were incomplete. Statistical issues. * Might have been contamination. * Narrative strategy that suggests a back-and-forth that teases out the meaning * Scientists do get passionate about things * And the back-and-forth (which is often part of science) makes the science better * Some interesting biases in how people present the data * Why choose an Ostrich when the sequence matches other things? * Why not use the hemoglobin match? * It *seems* that the result was correct * Ownership of data * Why didn't they reveal their data initially? * You can't publish things that don't work * Interesting question of what control and "junk data" are * Collisions between models * Lots of computer scientists like to share data * Biologists are used to making judgements about what gets released and what does not get released * Also the "ownership of stuff" thing * If you've created the data, and it might lead to more papers, shouldn't you get to keep it? * Or should we care more about the general advancement of knowledge * Running facilities costs a lot of money * A high impact publication means a LOT, and you don't always know what's in the data * Sam's pointless stories * "Significant difference" in learning outcomes. (If you do 20 measures, it doesn't mean much that one is significantly different.) * Dead sea scrolls and religious studies * What would be "extroardinary evidence"? Think about it as you do your paper reviews.