Project #1
Project 1: Implement suffix trees
Handed out: 2/26/08
Due: 3/25/08
For this project you will have to write software that builds a suffix tree from a DNA string (T)
provided in the input, outputs the suffix tree you constructed, and identifies the longest substring
of a second DNA string (P) that matches a substring of T.
Deliverables: Source code
Requirements:
1. The source code compiles and executes on a Linux machine (Mac OSX or cygwin under
Windows is OK). IMPORTANT: if your code doesn’t compile or run as “advertised” I
will not try to debug it and will consider the project failed.
2. The running time of the program is linear in size of input, both for constructing the tree
and for executing the search. By this, I mean that if I try several inputs of increasing
sizes, the running time grows proportionally to the size of the inputs (that’s what suffix
trees are supposedly good at :) ).
3. The output is correct (should go without saying).
4. You also provide me with a user manual (< 1 page) that explains how to install and run
your program.
5. The source code is readable: formatted nicely, and contains comments.
Grading: Maximum grade is 100 points. Each of requirements 2, 4, and 5 are worth 10 points.
Requirement 3 is worth 20 points (and includes both finding the match, and the correctness of
the tree output by the program). Failure to achieve requirement 1 results in an automatic score of
40 for this project.
Note: you can use any programming language that allows you to achieve these requirements.
Details
Input formats
The inputs will be provided in FASTA format (see http://en.wikipedia.org/wiki/Fasta_format)
with the following assumptions:
•The header line contains one single identifier immediately following the “>” character.
•The DNA sequence contains only characters A, C, T, and G (alphabet size = 4) but may
be in both lower and upper case.
•Each file (pattern and text) contains exactly one sequence.
Here is an example:
>Text
ACAGGTAGCAGGGAC
CATGACCAGGGCTGC
GAC
