Difference between revisions of "DOCK GA Development Goals"

From Rizzo_Lab
Jump to: navigation, search
(Created page with "==To Do List== -xover- guided based on score 1. good v good 2. bad v good 3. bad v bad nonexhaustive xover (pick subset of xover based on probability) 2-3 point xove...")
 
(To Do List)
(23 intermediate revisions by 2 users not shown)
Line 1: Line 1:
 +
{| border="1" cellpadding="8" cellspacing="0" style="background:white; text-align:left; width:90%"
 +
|- style="background:lightblue"
 +
! style="width:75%" !|Tasks
 +
! style="width:25%" !|src
 +
! style="width:13%" !|Owner
 +
! style="width:10%" !|Complete?
 +
|-
 +
|Fix bug that collapsed atom coordinates  || everywhere? nowhere? somewhere. || LEP || YES MUAHAHAHAHA!
 +
|-
 +
|Added Delimeter header ||conf_gen_ga || LEP || Yes
 +
|-
 +
|Fix xover only feature ||conf_gen_ga || LEP || Yes
 +
|-
 +
|Put in error messages for mut_rate > 1 || conf_gen_ga ||LEP || Yes
 +
|-
 +
|Manual user-defined mutation type ||conf_gen_ga || LEP || Yes
 +
|-
 +
|Remove check only option || conf_gen_ga || LEP || Yes
 +
|-
 +
|Add single molecule evolution in testcase in install dir. || install/test/genetic || LEP ||no
 +
|-
 +
|Compute delta slope of fitness score || congen_ga.cpp || LEP ||no
 +
|}
 +
<br>
 +
 +
 
==To Do List==
 
==To Do List==
-xover-
+
# slow down molecular evolution so there are less drastic canges between each successive generation
guided based on score  
+
# tanimoto coefficient percent change - might be inaccurate due to tan coef behavior
  1. good v good
+
# Rotatable bond changes
  2. bad v good
+
# Number of aromatic rings
  3. bad v bad
+
#-xover (guided based on score) - Good v Good ; Bad v Good ; Bad v Bad THIS
nonexhaustive xover (pick subset of xover based on probability)
+
#nonexhaustive xover (pick subset of xover based on probability)
2-3 point xover at once
+
# nonexhaustive xover for each pair of parents - have a set number of bonds for xover
adaptive maintenance ensemble based on ensemble convergence
+
#2-3 point xover at once
bring in new parents based on convergence
+
#adaptive maintenance ensemble based on ensemble convergence THIS
-Mutations-
+
#bring in new parents based on convergence
adaptive mutation rate
+
#Mutations-
pick location of mutation based on something
+
##adaptive mutation rate THIS
pick mutation type based on behavior of esemble
+
##pick location of mutation based on something
  too large boost deletion
+
##pick mutation type based on behavior of ensemble
  too small more add
+
##molecules too large boost deletion
  change ...boost replace/sub
+
##molecules too small, add more groups
mutation type selection based on probability vs ensemble
+
##change ...boost replace/sub
  complete x # y mutation so far so less prevalent etc
+
##mutation type selection based on probability vs ensemble
  3 layer subs do no work so don't do them
+
##complete x # y mutation so far so less prevalent etc
replace > 1 segment
+
##3 layer subs do no work so don't do them
-fitness-
+
##replace > 1 segment
turn on and off niching adaptive/extinction
+
#fitness-
reduce boost of fragments and all poor mols with niching
+
##turn on and off niching adaptive/extinction
pareto/mulitobjective ga
+
##reduce boost of fragments and all poor mols with niching
-selection-
+
##pareto/mulitobjective ga
metropolis selection for tournament/roulette
+
#selection-
adaptive keep #p and #o
+
##metropolis selection for tournament/roulette
-extinction-  
+
##adaptive keep #p and #o
automated vs on convergence
+
#extinction-  
-stop-
+
##user defined point vs on-the-fly convergence THIS
convergence
+
#stop-
-which molecules are best-
+
##convergence
-best first pruning - now uses descriptor score even if niching ned to delta to fitness/niching when used
+
#which molecules are best-
 +
##best first pruning - now uses descriptor score even if niching ned to delta to fitness/niching when used
 +
## geometric diversity using Hingarian and Tan pruning
 +
 
 +
== Known Bugs ==
 +
#Molecules Processed bug (dock.cpp)
 +
#verbose mol stats (amber typer)
 +
#molecule being renamed when going into repl even tho it's the same molecule

Revision as of 11:21, 20 May 2021

Tasks src Owner Complete?
Fix bug that collapsed atom coordinates everywhere? nowhere? somewhere. LEP YES MUAHAHAHAHA!
Added Delimeter header conf_gen_ga LEP Yes
Fix xover only feature conf_gen_ga LEP Yes
Put in error messages for mut_rate > 1 conf_gen_ga LEP Yes
Manual user-defined mutation type conf_gen_ga LEP Yes
Remove check only option conf_gen_ga LEP Yes
Add single molecule evolution in testcase in install dir. install/test/genetic LEP no
Compute delta slope of fitness score congen_ga.cpp LEP no



To Do List

  1. slow down molecular evolution so there are less drastic canges between each successive generation
  2. tanimoto coefficient percent change - might be inaccurate due to tan coef behavior
  3. Rotatable bond changes
  4. Number of aromatic rings
  5. -xover (guided based on score) - Good v Good ; Bad v Good ; Bad v Bad THIS
  6. nonexhaustive xover (pick subset of xover based on probability)
  7. nonexhaustive xover for each pair of parents - have a set number of bonds for xover
  8. 2-3 point xover at once
  9. adaptive maintenance ensemble based on ensemble convergence THIS
  10. bring in new parents based on convergence
  11. Mutations-
    1. adaptive mutation rate THIS
    2. pick location of mutation based on something
    3. pick mutation type based on behavior of ensemble
    4. molecules too large boost deletion
    5. molecules too small, add more groups
    6. change ...boost replace/sub
    7. mutation type selection based on probability vs ensemble
    8. complete x # y mutation so far so less prevalent etc
    9. 3 layer subs do no work so don't do them
    10. replace > 1 segment
  12. fitness-
    1. turn on and off niching adaptive/extinction
    2. reduce boost of fragments and all poor mols with niching
    3. pareto/mulitobjective ga
  13. selection-
    1. metropolis selection for tournament/roulette
    2. adaptive keep #p and #o
  14. extinction-
    1. user defined point vs on-the-fly convergence THIS
  15. stop-
    1. convergence
  16. which molecules are best-
    1. best first pruning - now uses descriptor score even if niching ned to delta to fitness/niching when used
    2. geometric diversity using Hingarian and Tan pruning

Known Bugs

  1. Molecules Processed bug (dock.cpp)
  2. verbose mol stats (amber typer)
  3. molecule being renamed when going into repl even tho it's the same molecule