|
| SoE home |
|
| Kevin Karplus's home page |
|
| Biomolecular Engineering Department |
| Date | Lecture Topic(s) | Due |
|---|---|---|
| Fri 26 Sept 2012 | administrivia, texts, assignments, structure of course | intake survey |
| Mon 1 Oct 2012 | Guest Lecture, Andrew Uzilov, Python tutorial | |
| Wed 3 Oct 2012 |
Discussion of UNIX permissions, chmod, how to submit programs.
What to comment in a program (meanings of variables, external view of
functions and whole program).
Why sequences are key to much of bioinformatics (low cost of DNA sequencing).
Error rates of information copying (polymerases):
DNA->DNA: Taq 1/125,000; Pfu 1/2,300,000; mitochondrial pol gamma ; 1/300,000-1/500,000; pol eta 1/18-1/38 RNA->DNA: HIV 1/1700-/8000; SIV 1/19,000; Accuscript RT 1/62,000 RNA->RNA: flu 1/10,000; Qbeta 1/3,000. Ribosome: Ecoli 1/3000; generally 1/2000-1/20,000. | |
| Fri 5 Oct 2012 | error rates and read lengths in sequencing. Notion of stochastic model as computable probability function. | Scaffolding |
| Mon 8 Oct 2012 | Review of returned homework: re.findall vs. re.split (keep re.compile outside loop), indent comments same as code, format() now preferred for Python, __doc__. Review definition of probability function, uniform distribution, i.i.d., event spaces, distribution over sequences of length n, lead up to length distribution * conditional distribution. | |
| Wed 10 Oct 2012 | Discussion of FASTA/FASTQ format details. Conditional probability, i.i.d. model, log probability, adding probabilities in log probability | |
| Fri 12 Oct 2012 | Mainly about fellowship applications | FASTA/FASTQ |
| Mon 15 Oct 2012 | Review of returned homework: suggestions about argparse (store_const and file opening), list comprehensions, izip_longest, translate. Markov chains: sum to 1 over strings of fixed length. Markov chains with stop: sum to 1 over all strings (left as homework). | |
| Wed 17 Oct 2012 | Markov chains with stop do sum to 1. Generating and parsing sequences with Markov model. Training Markov model. Higher-order Markov model. MLE and MAP (pseudocounts) estimates of model parameters. | |
| Fri 19 Oct 2012 | entropy, encoding cost, information gain. | fellowship |
| Mon 22 Oct 2012 | feedback on writing. Somewhat sloppy introduction to classifiers and converting P(data) stochastic models in to classifiers. | |
| Wed 24 Oct 2012 | Details on Markov chains. Using (scaled) lower-order Markov chains for pseudocounts. Started "Better than Chance" talk, digression into P-values and E-values. | |
| Fri 26 Oct 2012 | middle of "Better than Chance" talk. Log scales for histograms. Log normal distribution for lengths. Local vs. global composition for null model. | Markov chains |
| Mon 29 Oct 2012 | review of homework. Palindrome null as product of conditionally independent models of lower order. Contact prediction null (uniform versus conditioned on separation) | |
| Wed 31 Oct 2012 | Review of E-value and p-value. Markov's inequality proven. Arbitrary score function and null model converted to stochastic model. Substitution matrix as arbitrary score and as log( P_aligned(x,y)/(P(x)P(y))) | |
| Fri 2 Nov 2012 | Guest Lecture, Katie Fortney,
Science Library
| palindromes |
| Mon 5 Nov 2012 | Go over palindrome homework: efficient palindrome generation, citations, fellow students as audience for writing, handling wild cards in center of odd palindrome (SW vs. X, using sum of expected counts). Started alignments: alignment as pairing, as edit operations (and other views). | |
| Wed 7 Nov 2012 | some clarifications about null models, brief discussion of plotting and curve fitting. Alignment scoring, global vs. local, different gap cost models (# gaps, linear, affine, arbitrary increasing), alignment graph. | |
| Fri 9 Nov 2012 | dynamic programming and memoized algorithms. algorithm for 2D arbitrary gap cost O(m^2 n^2). algorithm for 1D arbitrary gap cost O(mn (m+n)). example for 1D algorithm. | null models |
| Mon 12 Nov 2012 | Veteran's Day, no class | |
| Wed 14 Nov 2012 | discussed homework returned. Looked at Pog contig graphs, for which I was not adequately prepared (I'd forgotten some of Bernick's notation). Referred students to Bernick's thesis. | |
| Fri 16 Nov 2012 | Cleaned up presentation of the Pog contig graphs. Had students help derive Needleman-Wunsch alighment. | |
| Mon 19 Nov 2012 | Further discussion of Pog contig graphs. Derivation by class of local alignment with linear gap cost. | |
| Wed 21 Nov 2012 | Derivation by students of Smith-Waterman. Pesentation of traceback algorithm. | research paper |
| Fri 23 Nov 2012 | Thanksgiving, no class | |
| Mon 26 Nov 2012 | re-presentation of traceback algorithm (example of global traceback). Intro to HMMs. | |
| Wed 28 Nov 2012 | return of research papers, feedback on writing, explanation of Bernick's computation of inversion requencies (cancellation of mapping biases), derivation of HMM forward algorithm | |
| Fri 30 Nov 2012 | affine-gap alignment | |
| Mon 3 Dec 2012 | Lagrangian Multipliers, Expectation Maximization, Backward algorithm, Baum-Welch training | |
| Wed 5 Dec 2012 | list of topic we could talk about. Genome assembly (read pairs and libraries, seqprep, k-mer cleanup, overlap graph) | |
| Fri 7 Dec 2012 | de Bruijn graph, k-mer counting tricks in Jellyfish (lock-free parallel hashing, packing key and count into one word) | degenerate codons |
| Wed 12 Dec 2012 | 8a.m.–11a.m. exam slot, not used |
|
|
| BME 205 home page | old BME 205 discussion forum | Fall 2012 BME 205 discussion forum | UCSC Bioinformatics research |
Questions about page content should be directed to
Kevin Karplus
Biomolecular Engineering
University of California, Santa Cruz
Santa Cruz, CA 95064
USA
karplus@soe.ucsc.edu
1-831-459-4250
318 Physical Sciences Building
