CS3824
Fall, 2014
Course Project

---

Project Description (PDF)

• Project Description (LaTeX)

• EmbedMotif.py

• RandomMotifFinder.py

• EvaluateLogLikelihood.py

---

Some Relevant Papers

• Timothy L. Bailey and Charles Elkan, "Unsupervised Learning of Multiple Motifs in Biopolymers Using Expectation Maximization". Machine Learning 21, 51-80 (1995).

• Leping Li, "GADEM: A Genetic Algorithm Guided Formation of Spaced Dyads Couple with an EM Algorithm for Motif Discovery". Journal of Computational Biology 16, 317-329 (2009).

---

Teams

Have been announced through the class listserv.

---

Test Sequences

1. sequences_k2_d0_n100_m10_mu0.fasta and EmbedMotif_k2_d0_n100_m10_mu0.log
   Created by

   EmbedMotif.py -k 2 -d 0 -n 100 -m 10 --mu 0 sequences_k2_d0_n100_m10_mu0.fasta
   

2. sequences_k12_d3_n10_m100_mu0.10.fasta and EmbedMotif_k12_d3_n10_m100_mu0.10.log
   Created by

   EmbedMotif.py -k 12 -d 3 -n 10 -m 100 --mu 0.10 sequences_k12_d3_n10_m100_mu0.10.fasta
   

3. sequences_k13_d5_n20_m100_mu0.15.fasta and EmbedMotif_k13_d5_n20_m100_mu0.15.log
   Created by

   EmbedMotif.py -k 13 -d 5 -n 20 -m 100 --mu 0.15 sequences_k13_d5_n20_m100_mu0.15.fasta
   

4. sequences_k10_d2_n20_m120_mu0.10.fasta and EmbedMotif_k10_d2_n20_m120_mu0.10.log
   Created by

   EmbedMotif.py -k 10 -d 2 -n 20 -m 120 --mu 0.10 sequences_k10_d2_n20_m120_mu0.10.fasta
   

---

Project Performance

Every motif finder, including RandomMotifFinder, was given 100 seconds of execution time.

Team	Test Sequence 4	Test Sequence 3
Alpha	8.4037	8.4085
Beta*	8.5245 (7.7386)	8.7568 (8.0933)
Gamma	13.3725	12.6686
Delta	13.9327	12.5671
Epsilon	11.6778	12.4178?
Optimal (Maybe?)	14.5021	13.5274
RandomMotifFinder	8.1303	8.3438

* Parallel Result (Sequential Result)

? Could not run EvaluateLogLikelihood on result

---