GRNmap Testing Report GRNmap-beta-bgf 2015-06-08 TM

From OpenWetWare

Jump to: navigation, search

Contents

Test Conditions

  • Date: 2015-06-08
  • Test Performed by: Tessa A. Morris, Electronic Notebook
  • Code Version: GRNmap-beta (updated 3:00 pm 2015-06-08; 9:11 am 2015-06-09; 9:29 am 2015-06-09)
  • MATLAB Version: 2014b
  • Computer on which the model was run: Row 2 #3 and 4

Purpose

  • The purpose is to check to make sure that the following changes have been implemented
1 optimization_diagnostics sheet with the requested output data.
a. added code functionality to compute min LSE and SSEs of individual genes.
2 reordering of the strain sigma sheets according to Issue #107
3 renamed graph files to correspond to gene names.
4 saved the diagnostic graph to a jpg.
5 fixed a bug in the penalty computation of the production rates.

Results

Fix b-0, P-0

 Operation terminated by user during
 general_least_squares_error (line 93)
 In finDiffEvalAndChkErr (line 25)
 In finitedifferences (line 127)
 In computeFinDiffGradAndJac (line 28)
 In barrier (line 573)
 In fmincon (line 799)
   [X,FVAL,EXITFLAG,OUTPUT,LAMBDA,GRAD,HESSIAN] =
   barrier(funfcn,X,A,B,Aeq,Beq,l,u,confcn,options.HessFcn,
   ...
 In lse (line 75)
       estimated_guesses =
       fmincon(@general_least_squares_error,estimated_guesses,[],[],[],[],lb,ub,[],options);
       
 In GRNmodel (line 32)
 GRNstruct = lse(GRNstruct);
Gene wt SSE dcin5 SSE
ACE2 1.57207E-08 3.14414E-08
AFT2 3.7395E-09 7.47901E-09
CIN5 1.07498E-07 1.07498E-07
FHL1 4.05182E-09 7.18506E-09
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 1.6000e-09
    • GRNstruct.GRNOutput.reg_out ans = 0.7776
 Operation terminated by user during general_least_squares_error (line 93)
 In finDiffEvalAndChkErr (line 25)
 In finitedifferences (line 127)
 In computeFinDiffGradAndJac (line 28)
 In barrier (line 573)
 In fmincon (line 799)
   [X,FVAL,EXITFLAG,OUTPUT,LAMBDA,GRAD,HESSIAN] =
   barrier(funfcn,X,A,B,Aeq,Beq,l,u,confcn,options.HessFcn, ...
 In lse (line 75)
       estimated_guesses =
       fmincon(@general_least_squares_error,estimated_guesses,[],[],[],[],lb,ub,[],options);
 In GRNmodel (line 32)
 GRNstruct = lse(GRNstruct);
Gene wt SSE dcin5 SSE
ACE2 1.49661E-08 2.99323E-08
AFT2 1.74801E-10 3.49602E-10
CIN5 1.07978E-07 1.07978E-07
FHL1 2.25952E-10 2.2613E-10
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 1.4426e-09
    • GRNstruct.GRNOutput.reg_out ans = 0.7778

Fix b-1, P-0

Gene wt SSE dcin5 SSE
ACE2 4.02469E-14 8.04939E-14
AFT2 2.42404E-14 4.84808E-14
CIN5 2.18625E-09 2.18625E-09
FHL1 3.40641E-12 4.57846E-12
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 2.2822e-11
    • GRNstruct.GRNOutput.reg_out ans = 0.9857
Gene wt SSE dcin5 SSE
ACE2 1.65905E-09 3.3181E-09
AFT2 2.35952E-09 4.71904E-09
CIN5 2.2864E-09 2.2864E-09
FHL1 2.03414E-09 2.14703E-09
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 1.2990e-10
    • GRNstruct.GRNOutput.reg_out ans = 0.9856
wt SSE |dcin5 SSE
4.02026E-14 |8.04052E-14
2.42757E-14 |4.85514E-14
2.18625E-09 |2.18625E-09
3.40659E-12 |4.57858E-12
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 2.2822e-11
    • GRNstruct.GRNOutput.reg_out ans = 0.9857
Gene wt SSE dcin5 SSE
ACE2 4.02026E-14 8.04052E-14
AFT2 2.42757E-14 4.85514E-14
CIN5 2.18625E-09 2.18625E-09
FHL1 3.40659E-12 4.57858E-12
  • Typing in the codes for LSE and penalty on MATLAB
    • GRNstruct.GRNOutput.lse_out ans = 2.2822e-11
    • GRNstruct.GRNOutput.reg_out ans = 0.9857

Discussion

Fix b-0, P-0

  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-0_fixP-0_graph (TolX and TolFun values set to 1e-8) did not work
  • For 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-0_fixP-0_graph with TolX and TolFun values set to 1e-6 an optimization_diagnostics sheet was output with the following information: LSE, Penalty, min LSE, iteration count, and wt SSE and dcin5 SSE for each gene
    • Order of sheets was wt_log2_optimized_expression_dcin5_log2_optimized_expression, wt_sigmas, dcin5_sigmas, optimized_production_rates, optimized_threshold_b, network_optimized_weights, optimization_diagnostics
    • Graph files were saved correspond to gene names
    • Automatically saved the diagnostic graph to a .jpg (named OptimizationDiagnostic.jpg)
  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-0_fixP-0_graph (MaxIter and MaxFunEval to 1e6) did not work
  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-0_fixP-0_graph (MaxIter and MaxFunEval to 1e5) output an optimization_diagnostics sheet with the following information: LSE, Penalty, min LSE, iteration count, and wt SSE and dcin5 SSE for each gene
    • Order of sheets was wt_log2_optimized_expression_dcin5_log2_optimized_expression, wt_sigmas, dcin5_sigmas, optimized_production_rates, optimized_threshold_b, network_optimized_weights, optimization_diagnostics
    • Graph files were saved correspond to gene names
    • Automatically saved the diagnostic graph to a .jpg (named OptimizationDiagnostic.jpg)

Fix b-1, P-0

  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-1_fixP-0_graph (TolX and TolFun values set to 1e-8)
  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-1_fixP-0_graph (TolX and TolFun values set to 1e-6)
  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-1_fixP-0_graph (MaxIter and MaxFunEval to 1e6)
  • 4-genes_6-edges_artificial-data_Sigmoidal_estimation_fixb-1_fixP-0_graph (MaxIter and MaxFunEval to 1e5)
Personal tools