MIT6_047f08_lec03_slide03

MIT6_047f08_lec03_slide03 - MIT OpenCourseWare...

Info icon This preview shows pages 1–7. Sign up to view the full content.

View Full Document Right Arrow Icon
MIT OpenCourseWare http://ocw.mit.edu 6.047 / 6.878 Computational Biology: Genomes, Networks, Evolution Fall 2008 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms .
Image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Rapid sequence alignment and Database search Local alignment, varying gap penalties Karp-Rabin: Semi-numerical methods BLAST: dB search, neighborhood search Statistics of alignment scores (recitation) Lecture 3 Thursday Sept 11, 2008 6.047/6.878 - Computational Biology: Genomes, Networks, Evolution
Image of page 2
DNA Genome Assembly Gene expression analysis Cluster discovery Gibbs sampling Protein network analysis Emerging network properties Regulatory network inference Challenges in Computational Biology 1 Gene Finding 5 Regulatory motif discovery Database search 3 Sequence alignment Evolutionary Theory 7 T C ATG C TAT T CG TGATA A TGA G GATAT T T AT C ATAT T T ATGAT T T Comparative Genomics 6 2 4 8 RNA transcript 9 10 11 13 12
Image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Tues: Sequence alignment + dynamic programming A C G T C A T C A A C G T G A T C A mutation A G T G T C A A G T G T C A deletion A G T G T C A T begin end A G T G T C A T insertion The sequence alignment problem Genomes change: mutation, insertions, deletions – Alignment: infer evolutionary events Scoring metric reflects evolutionary properties AGTGCCCTGGAACCCTGACGGTGGGTCACAAAACTTCTGGA AGTGACCTGGGAAGACCCTGACCCTGGGTCACAAAACTC Needleman-Wunsch algorithm Local update rule: F(i,j) = max{up, left, diagonal} Save choice pointers for traceback Bottom-right corner gives optimal alignment score Trace-back of pointers gives optimal path/alignment A C G T C A T C A T A G T G T C A A G T C/G T C A Dynamic programming and sequence alignment Alignment scores are additive: decomposable Represent sub-problem scores in M(i,j) matrix Duality between alignment and path through matrix Dynamic programming Problems that can be decomposed into subparts Identical sub-problems: reuse computation Bottom-up approach: systematically fill table
Image of page 4
Today’s Goal: Diving deeper into alignments 1. Global alignment vs. Local alignment Needleman-Wunsch and Smith-Waterman Varying gap penalties and algorithmic speedups 2. Linear-time exact string matching Karp-Rabin algorithm and semi-numerical methods Hash functions and randomized algorithms 3. The BLAST algorithm and inexact matching Hashing with neighborhood search Two-hit blast and hashing with combs 4. Probabilistic foundations of sequence alignment Mismatch penalties, BLOSUM and PAM matrices Statistical significance of an alignment score
Image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full Document Right Arrow Icon
Today’s Goal: Diving deeper into alignments 1. Global alignment vs. Local alignment Needleman-Wunsch and Smith-Waterman Varying gap penalties and algorithmic speedups 2. Linear-time exact string matching Karp-Rabin algorithm and semi-numerical methods Hash functions and randomized algorithms 3. The BLAST algorithm and inexact matching Hashing with neighborhood search Two-hit blast and hashing with combs 4.
Image of page 6
Image of page 7
This is the end of the preview. Sign up to access the rest of the document.

{[ snackBarMessage ]}

What students are saying

  • Left Quote Icon

    As a current student on this bumpy collegiate pathway, I stumbled upon Course Hero, where I can find study resources for nearly all my courses, get online help from tutors 24/7, and even share my old projects, papers, and lecture notes with other students.

    Student Picture

    Kiran Temple University Fox School of Business ‘17, Course Hero Intern

  • Left Quote Icon

    I cannot even describe how much Course Hero helped me this summer. It’s truly become something I can always rely on and help me. In the end, I was not only able to survive summer classes, but I was able to thrive thanks to Course Hero.

    Student Picture

    Dana University of Pennsylvania ‘17, Course Hero Intern

  • Left Quote Icon

    The ability to access any university’s resources through Course Hero proved invaluable in my case. I was behind on Tulane coursework and actually used UCLA’s materials to help me move forward and get everything together on time.

    Student Picture

    Jill Tulane University ‘16, Course Hero Intern