Basic Tutorial: Difference between revisions

From DISI
Jump to navigation Jump to search
No edit summary
 
No edit summary
 
(8 intermediate revisions by 4 users not shown)
Line 1: Line 1:
To run SEA we need three query files:
See [[SEA]]
* smiles : The separator between smiles and IDs is a semicolumn
 
* fingerprints : Separator is also a semicolumn. Can be generated by the following command:
To run [[SEA]] we need, for both the query and the reference, three files with the same basename and different extensions:  
* <b>smiles</b> (.smi) : The separator between smiles and IDs is a semicolon. Make sure it has the same characteristics than the reference by running the following command:
  > sea-molecule-clean
* <b>fingerprints</b> (.fp) : Separator is also a semicolon. Can be generated by the following command:
   > sea-molecule-fingerprint
   > sea-molecule-fingerprint
* set file : First element is the set code, the second is the set description, the third is the list of compound IDs in the set separated by a coumn.
* <b>sets</b> (.set) : First element (separated by semicolon) is the set code, the second is the set description, the third is the list of compound IDs in the set separated by a colon.
 
Additionally, we will need the model fit file generated from [[SEA]].
 
The actual command to <b>run SEA</b> is:
 
  > sea-run -f model_fit_file model_set_file query_set_file
 
Export the results
 
  > sea-export -t 0.5 -o sea_output 9
 




For the reference sets we need the same three files plus the fits file, generated from the
[[Category:SEA]]
[[Category:Tutorials]]
[[Category:Internal]]

Latest revision as of 06:46, 13 March 2014

See SEA

To run SEA we need, for both the query and the reference, three files with the same basename and different extensions:

  • smiles (.smi) : The separator between smiles and IDs is a semicolon. Make sure it has the same characteristics than the reference by running the following command:
  > sea-molecule-clean
  • fingerprints (.fp) : Separator is also a semicolon. Can be generated by the following command:
  > sea-molecule-fingerprint
  • sets (.set) : First element (separated by semicolon) is the set code, the second is the set description, the third is the list of compound IDs in the set separated by a colon.

Additionally, we will need the model fit file generated from SEA.

The actual command to run SEA is:

 > sea-run -f model_fit_file model_set_file query_set_file

Export the results

 > sea-export -t 0.5 -o sea_output 9