?

Log in

No account? Create an account

Previous Entry | Next Entry

BLAST and dynamic programming r 2 1337

As those of you who have been following the story of my course, CIS 690, Data Mining in Bioinformatics, may recall, I have changed the prerequisite from CIS 300 (Data Structures and Algorithms) to CIS 200 (Introduction to Computer Science).

There are 10 people in the class but only 5 are taking it: 3 CIS undergrads, 1 Ph.D. student in Math, and 1 Ph.D. student in Statistics. That leaves 5 auditors: 1 M.S. student in CIS (and Ph.D. student in Stats), 1 Ph.D. student in CIS, 1 statistics faculty member, and 2 SUROP students (Math and CS).

"How does one cater to such a diverse audience?" I wondered. A couple of them are quite familiar with BLAST and dynamic programming, while the whole concept of string matching algorithms (or algorithms in general) is new to others.

Thus did I come up with the following example for explaining edit distance and k-approximate string matching today:

--u-r- 2-- q--l 4-- mee!1!!
++|+|+ x++ x++| x++ ||-|---    insert (10) / delete (4) / twiddle (3)
you're too cool for me-!---


History is philosophy teaching by examples.
    -Thucydides

kewl
xxx|                           twiddle (3)
cool

1337
xxxx                           twiddle (4) or twiddle (0) // BLASTx?
leet

--ur
++||                           insert (2)
your


You know, I think that middle one might be a good way to explain BLASTx.

--
Banazir

Comments

( 5 comments — Leave a comment )
crypthanatopsis
Jun. 20th, 2005 06:19 am (UTC)
I believe the technical term is 'Levenshtein distance', is it not?
grain_king
Jun. 20th, 2005 06:41 am (UTC)
gesundheit!
banazir
Jun. 21st, 2005 03:47 am (UTC)
Edit distance
Perhaps. I prefer "normal" to "Gaussian" even though I use Gaussian. Only for terms that are known better as proper nouns (Jacobian, Laplacian, Hamiltonian tour) will I prefer the proper noun.

Do you know, for the first 3 months in 1990 after I first heard "edit distance" (from Steve Salzberg, my undergrad academic advisor), I thought it was "Eta distance"?

--
Banazir
casecob
Jun. 20th, 2005 11:25 am (UTC)
Speaking of bioinfo, right now i'm working on a few multiple sequence alignments.

I hate the ClustalW algorithm.

Make me something better ;)
banazir
Jun. 21st, 2005 03:48 am (UTC)
Ya had to remind me
I need to cover ClustalW later this week.

Today was Bayesian network structure learning, but I got off on digressions into Bayes optimal classification and EM again.

I really need to rewrite the lecture to cover Structural EM, is what.

--
Banazir
( 5 comments — Leave a comment )

Latest Month

December 2008
S M T W T F S
 123456
78910111213
14151617181920
21222324252627
28293031   

KSU Genetic and Evolutionary Computation (GEC) Lab

Teunciness

Breakfast

Science, Technology, Engineering, Math (STEM) Communities

Fresh Pages

Tags

Powered by LiveJournal.com
Designed by Naoto Kishi