NTNU NTNU-Medisin
   


 MAIN

 RESEARCH

 MEMBERS

 PUBLICATIONS

 SERVICES

 COURSES

 CONFERENCES

 EMPLOYMENT

 JOURNALS

 MEETINGS

 CONTACT


 TOOLS
  Single motif
  Composite motif
  Promoter fetcher
  Compo
  genePI
  PriorsEditor

 DATASETS



 LOGIN

 Last Update:
  September 27, 2017




Composite motif benchmark



Step 1: Register as a user and login.

Step 2: Download test sequences and PWMs. TRANSCompel contains 20 directories with random selected decoy matrices for 99,95,90,75,50 and 0 % noise. 10 for custom motifs and 10 for transfac motifs. For Liver and Muscle, only fasta sequences and pwms are available.
TRANSCompel (tar.gz) (zip)
Liver (tar.gz) (zip)
Muscle (tar.gz) (zip)

Step 3: Run your method on the downloaded dataset.

Step 4: Parse your results to our predefined result format.
Example: AP1-Ets_pred.txt

Step 5:
TRANSCompel:Put your results from each run (10 sequences) in predefined subdirectories under predictions. i.e. custom_0/noise_0 results should be put in: yourmethod/yourmethod_custom_0_noise_0/AP1-Ets_pred.txt

Muscle/Liver:Organize your result like this: yourmethod/muscle_pred.txt

Step 6: Upload prediction directory. File should be a tar.gz file. Filename must consist of the name of the sequence ending with _pred.txt.
Example: yourmethod/yourmethod_custom_0_noise_0/AP1-Ets_pred.txt
Example: yourmethod/yourmethod_custom_0_noise_0/AP1-NFAT_pred.txt

Step 7: Run evaluation




View article results:

Plot Standard Deviation on dataset plot
Choose dataset source:
Choose performance measure: CC: Sn: SP:
PPV: PC: ASP:




Bioinformatics & Gene Regulation
Department of Cancer Research and Molecular Medicine
Norwegian University of Science and Technology
Trondheim, Norway