Here is an example of a 3-state fit. The data files are
3state.01.asc and 3state.02.asc . The parameters used to synthesize the data are in
3state.true.par. Please see the file tutorial.txt for an older but (relatively) more complete tutorial. The aim for this tutorial is to read in the two denaturation curves and fit them globally to a three state denaturation model. Characters that the user enters are in
bold. Program output/feedback is in regular font. My added comments to this log file are in
italics.Some quick notes:
- All Savuka commands consist of a two or three letter command followed by switches and/or input parameters.
- The command switch (much like those in Unix) must be entered on the command line (contains the prompt “SA>”)
- The input parameters may be entered on the command line also. If parameters are not given on the command line, the program will prompt the user for all parameters required by the command and optional switches. In other words, if you are not sure what the input arguments are for a command you can simply hit return and expect to be prompted. If you are unsure of which switch to use, consult the help manual.
- The latest version of the help manual is online and may be accessed by entering help or help command).
- Savuka uses PGPLOT and GrWin (on Windows) for its graphic engine. It can also handle X11 on Unix machines.
protein 12% savuka
SAVUKA Version 5.2.0
7/16/99 Irix/Linux
Copyright(C) 1991, 1996
David G. Lambright and Osman Bilsel
Enter: "help" to: get info about commands
"exit" quit Savuka
"clear -s" re-dimension work space
"!" shell to system
Questions? E-mail osman@protein.chem.psu.edu
MAX # OF BUFFERS= 256, MAX # OF POINTS= 2048
SA> rea 3state.01.asc # Read the data file 3state.01.asc. Format is ASCII xy pairs.
FIRST LINE TO READ = [ 1 ] # The absence of an entry indicates that I accepted the default (1).
NUMBER OF LINES TO READ = [ 2048 ]
STORE EVERY NTH LINE = [ 1 ]
ROW VALUE ASSOCIATED WITH COLUMN = [ 0.0000000E+00 ] # The row-value is for multidimensional data (up to 5-dimensions).
SA> rea 3state.02.asc
FIRST LINE TO READ = [ 1 ]
NUMBER OF LINES TO READ = [ 2048 ]
STORE EVERY NTH LINE = [ 1 ]
ROW VALUE ASSOCIATED WITH COLUMN = [ 0.0000000E+00 ]
SA> dir # The data is placed in buffers (similar to columns in a spreadsheet)
BUFFER #POINTS ROW-VALUE REMARKS
------ ------- --------- --------------------------------
1 50 0.0000E+00 3state.01.asc
2 50 0.0000E+00 3state.02.asc
SA> pl # define the active region (similar to selecting a block in a spreadsheet).
FIRST BUFFER = [ 1 ]
LAST BUFFER = [ 2 ]
FIRST POINT = [ 1 ]
LAST POINT = [ 128 ] 50
NOTE: USE "EXP" TO ACTIVATE PLOT LIMITS
SA> dr 1 2 # display the data graphically on the screen. This command overlays all data on one plot.
![]()
SA> fun # select a function to fit to the data.
FUN> DEFINE A MODEL FUNCTION BY SELECTING ANY
FUN> SEQUENCE OF THE FOLLOWING SUB-FUNCTIONS.
Generic functions:
CONSTANT (ADDITIVE) ........................1
CONSTANT (MULTIPLICATIVE) .................30
ZERO-TIME SHIFT ...........................26
LINEAR ....................................27
QUADRATIC .................................28
EXPONENTIAL ................................2
STRETCHED EXPONENTIAL ......................3
GAUSSIAN ...................................4
LORENTZIAN ................................40
SIN (A*sin(b*pi*x + d) ..................80
STERN-VOLMER (F/Fo) .......................55
Exponentials convolved with Gaussians:
CONSTANT W/ PULSE, t<to ....................5
CONSTANT W/ PULSE, t>to ....................6
EXPONENTIAL W/ PULSE .......................7
ENTER <CR> FOR MORE, "Q" TO QUIT DISPLAY:
Kinetic models (w/o denaturant dependence):
2A -> B -> C ...............................8
2A -> 2B -> C ..............................9
U <-> I <-> N .............................46
U <-> I1 <-> I2 <-> N .....................24
I1 <-> I2 <-> I3 <-> I4 <-> I5 ............44
A + B -> C -> D ..........................34
A + B -> C -> D with k-1 -> Kd ...........68
2A -> B -> C with k-1 -> Keq..............51
Kinetic models (with denaturant dependence):
2A -> B -> C ..............................10
2U -> I -> N -> 2U ........................18
2U <=> 2I <-> N <-> 2U ....................64
2U <=> I <-> N <-> 2U .....................71
2U -> 2I -> N -> 2U .......................19
U <-> I <-> N .............................47
U <-> I1 <-> I2 <-> N .....................25
I1 <-> I2 <-> I3 <-> I4 <-> I5 ............45
ENTER <CR> FOR MORE, "Q" TO QUIT DISPLAY:
Combined kinetic/equil models (w/ denat):
U <-> I1 <-> I2* <-> I3 <-> N .............54
U* <-> I1* <-> I2* <-> I3 <-> N ...........56
U* <-> I1* <-> I2* <-> I3 <-> N (fix conc).70
(U <-> I1)abc <-> N .......................66
{U}abc <-> {I1}abc <-> N ..................67
{U}abc <-> {I1}abc <-> {N}abc .............75
{U}abc <-> {I1}abc <-> {N}abc .............77
{U}abc <-> {I1}abc <-> {I2}abc <-> {N}abc..74
{U}abcd <-> {I1}abcd <-> {I2}abcd<->{N}bc..84
{U}abcd <-> {I1}abcd <-> {I2}abcd<->{N}ab..86
{U}abcd <-> {I1}abcd <-> {I2}abcd<->{N}bc..91
2U <-> N ..................................97
Kinetic models (thermal):
A -> B ....................................52
2A -> B -> C ..............................83
Kinetic models (denaturant + thermal):
A -> B (Tanford denat binding).............53
A -> B (linear).,,,,,,,,,,,,,,,............82
2U <-> N ..................................96
Equilibrium unfolding (denaturant):
N <-> U ..................................14
N <-> I <-> U (w/ Z-value approx.).........11
N <-> I <-> U (w/o Z-value approx.)........88
N <-> I1 <-> I2 <-> U .....................21
N <-> I1 <-> I2 <-> U (w/o Z-value approx.)89
N <-> I1 (<-> D) <-> I2 <-> U .............79
N <-> 2U ..................................15
N <-> I <-> 2U ............................16
N <-> 2I<-> 2U ............................69
N <-> 3U .................................41
N3X3 <-> 3U + 3X .........................72
N <-> 4U ........ .........................93
N <-> 2I <-> 4U ...........................95
Equilibrium unfolding (denaturant;anisotropy):
N <-> U ..................................48
N <-> I <-> U .............................49
N <-> I1 <-> I2 <-> U .....................85
N <-> I <-> U (w/o Z-value approx.)........92
N <-> I1 <-> I2 <-> I3 <-> U ..............87
Equilibrium unfolding (thermal):
N -> U ....................................35
N -> I -> U ...............................36
N -> 2U ...................................37
N -> I -> 2U ..............................38
N -> 2I -> 2U .............................39
N <-> 3U .................................43
Equilibrium unfolding (thermal+denaturant):
N -> 2U ...................................94
ENTER <CR> FOR MORE, "Q" TO QUIT DISPLAY:
Ultracentrifugation:
Eq sedimentation single ideal species .....59
Eq sedimentation single non-ideal species..60
Monomer-dimer eq sedimentation ............57
Monomer-dimer-trimer-tetramer eq sediment..58
Custom models:
Trp Repressor kinetic model 1 .............12
Trp Repressor kinetic model 2 .............13
Ras P21-ternary complex 2-state eq unfold..42
Ras P21-binary complex 2-state eq unfold..61
Ras P21+APOP21 complex 2-state eq unfold...62
2M + nH <-> MHn <-> DHn ...................33
Exponential w/ denaturant dependence.......22
N <-> I <-> 2U KIN->EQ ...................17
I1 <-> I2 <-> I3 <-> I4 <-> I5 KIN->EQ....50
2-Site binding model ......................76
Glorias 2-active form w 3-ionizable grps ..78
Time-resolved anisotropy models:
Ivv or Ivh, EXPONENTIAL ...................20
Ivv or Ivh, GLOBALIZED A -> B .............23
Ivv or Ivh, GLOBALIZED A -> B -> C ........29
Ivv or Ivh, GLOBALIZED {A} -> {B} -> C.....90
Ivv or Ivh, GLOBALIZED B <-> A -> C .......31
Ivv or Ivh, ELLIPSOID .....................32
SUB-FUNCTION # (ENTER "0" TO END) = [ 0 ] 11 # function 11 is the 3-state equilibrium model
SUB-FUNCTION # (ENTER "0" TO END) = [ 0 ]
SA> lp # list parameters of the model
LP> ENTER "0" WHEN DONE
LP> ENTER BUFFER NUMBER = [ 1 ] # each parameter is described by a buffer number and parameter number
BUFFER 1 [ 0.0000000E+00] 3state.01.asc # each parameter also has a link, either to itself or another parameter
1 Gni (Kcal/mol) 0.000000 1 1 VAR
2 Ani (Kcal/mol/[D]) 0.000000 1 2 VAR
3 Giu (Kcal/mol) 0.000000 1 3 VAR
4 Aiu (Kcal/mol/[D]) 0.000000 1 4 VAR
5 SLOPE OF NATIVE BASELINE 0.000000 1 5 VAR
6 Y-INT OF NATIVE BASELINE 0.000000 1 6 VAR
7 SLOPE OF UNFOLDED BASELIN 0.000000 1 7 VAR
8 Y-INT OF UNFOLDED BASELIN 0.000000 1 8 VAR
9 Z PARAMETER 0.000000 1 9 VAR
10 TEMPERATURE (K) 0.000000 1 10 FIXED
LP> ENTER "0" WHEN DONE
LP> ENTER BUFFER NUMBER = [ 2 ]
BUFFER 2 [ 0.0000000E+00] 3state.02.asc # buffer parameter var/fixed
1 Gni (Kcal/mol) 0.000000 1 1 VAR # note that this parameter is linked to Gni of buff 1
2 Ani (Kcal/mol/[D]) 0.000000 1 2 VAR # " " " " " " " Ani " " "
3 Giu (Kcal/mol) 0.000000 1 3 VAR # etc.
4 Aiu (Kcal/mol/[D]) 0.000000 1 4 VAR
5 SLOPE OF NATIVE BASELINE 0.000000 2 5 VAR # by default, optical properties are not global.
6 Y-INT OF NATIVE BASELINE 0.000000 2 6 VAR
7 SLOPE OF UNFOLDED BASELIN 0.000000 2 7 VAR
8 Y-INT OF UNFOLDED BASELIN 0.000000 2 8 VAR
9 Z PARAMETER 0.000000 2 9 VAR
10 TEMPERATURE (K) 0.000000 1 10 FIXED # temperature is global and fixed.
LP> ENTER "0" WHEN DONE
LP> ENTER BUFFER NUMBER = [ 3 ] 0
SA> ap # alter/adjust parameters, allows you to enter parameter values/initial guesses.
AP> ENTER "0" WHEN DONE
AP> ENTER BUFFER NUMBER = [ 1 ]
AP> BUFFER NUMBER 1 [ 0.0000000E+00]
AP> <CR> TO RETAIN CURRENT VALUES
Gni (Kcal/mol) = [ 0.0000000E+00 ] 4
Ani (Kcal/mol/[D]) = [ 0.0000000E+00 ] 1
Giu (Kcal/mol) = [ 0.0000000E+00 ] 7
Aiu (Kcal/mol/[D]) = [ 0.0000000E+00 ] 2
SLOPE OF NATIVE BASELINE = [ 0.0000000E+00 ]
Y-INT OF NATIVE BASELINE = [ 0.0000000E+00 ]
SLOPE OF UNFOLDED BASELINE = [ 0.0000000E+00 ]
Y-INT OF UNFOLDED BASELINE = [ 0.0000000E+00 ] 11
Z PARAMETER = [ 0.0000000E+00 ] .6
TEMPERATURE (K) = [ 0.0000000E+00 ] 298.15
AP> ENTER "0" WHEN DONE
AP> ENTER BUFFER NUMBER = [ 2 ]
AP> BUFFER NUMBER 2 [ 0.0000000E+00]
AP> <CR> TO RETAIN CURRENT VALUES
Gni (Kcal/mol) = [ 4.0000000 ] # note how the linked parameters are automatically updated: it's the same parameter
Ani (Kcal/mol/[D]) = [ 1.0000000 ]
Giu (Kcal/mol) = [ 7.0000000 ]
Aiu (Kcal/mol/[D]) = [ 2.0000000 ]
SLOPE OF NATIVE BASELINE = [ 0.0000000E+00 ]
Y-INT OF NATIVE BASELINE = [ 0.0000000E+00 ]
SLOPE OF UNFOLDED BASELINE = [ 0.0000000E+00 ]
Y-INT OF UNFOLDED BASELINE = [ 0.0000000E+00 ] 11
Z PARAMETER = [ 0.0000000E+00 ] .3
TEMPERATURE (K) = [ 298.1500000 ]
AP> ENTER "0" WHEN DONE
AP> ENTER BUFFER NUMBER = [ 3 ] 0
SA> mod # calculate the model function
SA> sd 1 # change the plotting so that the data and model are superimposed. (How good are the initial guesses?)
SA> dr 1 2
SA> rms -p 1 -last 10 # let's estimate the noise level in the data from the last 10 points. -p 1 indicates linear.
FIRST BUFFER = [ 1 ]
LAST BUFFER = [ 2 ]
POLYNOMIAL COEFICIENTS
SA> C(0) = 0.9121461E+01
SA> C(1) = -0.8887876E-01
BUFFER= 1 RMS DEVIATION= 0.0170229 AVERAGE= 0.0000000
POLYNOMIAL COEFICIENTS
SA> C(0) = 0.9092176E+01
SA> C(1) = -0.1034156E+00
BUFFER= 2 RMS DEVIATION= 0.0326645 AVERAGE= 0.0000000
SA> fit # all set to start a least squares fit. This uses the Marquardt algorithm.
ITERATION LIMIT = [ 10 ]
ITERATIONS= 0, RCHISQ= 19952.8249714
ITERATIONS= 1, RCHISQ= 1194.4691027, %CHANGE= 94.01353
ITERATIONS= 2, RCHISQ= 1025.1202235, %CHANGE= 14.17775
ITERATIONS= 3, RCHISQ= 899.5766963, %CHANGE= 12.24671
ITERATIONS= 4, RCHISQ= 813.7570586, %CHANGE= 9.54000
ITERATIONS= 5, RCHISQ= 758.0014120, %CHANGE= 6.85163
ITERATIONS= 6, RCHISQ= 654.9079986, %CHANGE= 13.60069
ITERATIONS= 7, RCHISQ= 513.8801550, %CHANGE= 21.53399
ITERATIONS= 8, RCHISQ= 361.4956974, %CHANGE= 29.65370
ITERATIONS= 9, RCHISQ= 190.5288375, %CHANGE= 47.29430
ITERATIONS= 10, RCHISQ= 93.3173257, %CHANGE= 51.02194
%-----------FIT INFO AND STATISTICS-----------%
DATA FROM FILE 3state.01.asc
RANGE OF POINTS USED FOR FIT: 1 50
MARQUARDT NON-LINEAR LEAST-SQUARES FIT
NUMERICAL PARTIAL DERIVATIVES
MATRIX INVERSION BY GAUSS-JORDAN
CORRESPONDING DATA RANGE: 0.0000, 9.0000
ITERATION LIMIT REACHED BEFORE CONVERGENCE
FIT TOLERANCE = 1.0000000000000000E-04
TOTAL # OF DATA POINTS = 100
TOTAL # OF VAR PARAMETERS = 14
NO. DEGREES OF FREEDOM = 86
CHI-SQUARED= 8025.290006970623
REDUCED CHI-SQUARED = 93.31732566244911
GOODNESS OF FIT = 0.0000000000000000E+00
%--------------- PARAMETERS ---------------%
value stddev dependency
>> BUFFER 1 [ 0.00000E+00] 3state.01.asc
Gni (Kcal/mol) = 3.0393096 +/- 0.2655361 0.9514315
----- ENTER <CR> TO CONTINUE OR Q TO QUIT ----- :
Ani (Kcal/mol/[D]) = 1.5838682 +/- 0.1114942 0.9418988
Giu (Kcal/mol) = 7.2939581 +/- 1.3015531 0.9625408
Aiu (Kcal/mol/[D]) = 1.7726415 +/- 0.3088753 0.9619133
SLOPE OF NATIVE BASELINE = -0.8172158 +/- 0.2879918 0.8787121
Y-INT OF NATIVE BASELINE = 1.2169147 +/- 0.1122522 0.6606653
SLOPE OF UNFOLDED BASELIN = -0.0107279 +/- 0.0281425 0.8862934
Y-INT OF UNFOLDED BASELIN = 9.3120813 +/- 0.1996915 0.9021817
Z PARAMETER = 0.9160288 +/- 0.0206363 0.8134253
TEMPERATURE (K) = 298.1500000
>> BUFFER 2 [ 0.00000E+00] 3state.02.asc
Gni (Kcal/mol) = 3.0393096 +/- 0.2655361 0.9514315
Ani (Kcal/mol/[D]) = 1.5838682 +/- 0.1114942 0.9418988
----- ENTER <CR> TO CONTINUE OR Q TO QUIT ----- :
Giu (Kcal/mol) = 7.2939581 +/- 1.3015531 0.9625408
Aiu (Kcal/mol/[D]) = 1.7726415 +/- 0.3088753 0.9619133
SLOPE OF NATIVE BASELINE = -0.5685914 +/- 0.2963942 0.8670376
Y-INT OF NATIVE BASELINE = 1.1900560 +/- 0.1986495 0.6804628
SLOPE OF UNFOLDED BASELIN = 0.0790031 +/- 0.0652208 0.9034655
Y-INT OF UNFOLDED BASELIN = 8.6038054 +/- 0.4773007 0.9142960
Z PARAMETER = 0.5522232 +/- 0.0559888 0.8516585
TEMPERATURE (K) = 298.1500000
FILE FOR STATISTICS:
FILE ALREADY EXISTS, OVERWRITE? [N]:
SA> fit # didn't converge. Let's try some more iterations.
ITERATION LIMIT = [ 10 ]
ITERATIONS= 0, RCHISQ= 93.3173257
ITERATIONS= 1, RCHISQ= 61.5332220, %CHANGE= 34.06024
ITERATIONS= 2, RCHISQ= 42.5853628, %CHANGE= 30.79289
ITERATIONS= 3, RCHISQ= 26.1556747, %CHANGE= 38.58060
ITERATIONS= 4, RCHISQ= 1.6750747, %CHANGE= 93.59575
ITERATIONS= 5, RCHISQ= 0.9804490, %CHANGE= 41.46834
ITERATIONS= 6, RCHISQ= 0.9803469, %CHANGE= 0.01042
ITERATIONS= 7, RCHISQ= 0.9803469, %CHANGE= 0.00000
%-----------FIT INFO AND STATISTICS-----------%
DATA FROM FILE 3state.01.asc
RANGE OF POINTS USED FOR FIT: 1 50
MARQUARDT NON-LINEAR LEAST-SQUARES FIT
NUMERICAL PARTIAL DERIVATIVES
MATRIX INVERSION BY GAUSS-JORDAN
CORRESPONDING DATA RANGE: 0.0000, 9.0000
CHISQ CONVERGED IN 7 ITERATIONS
FIT TOLERANCE = 1.0000000000000000E-04
TOTAL # OF DATA POINTS = 100
TOTAL # OF VAR PARAMETERS = 14
NO. DEGREES OF FREEDOM = 86
CHI-SQUARED= 84.30983142524234
REDUCED CHI-SQUARED = 0.9803468770377016 # This is close to 1, a pretty good fit.
GOODNESS OF FIT = 0.5313977911240007
%--------------- PARAMETERS ---------------%
value stddev dependency
>> BUFFER 1 [ 0.00000E+00] 3state.01.asc
Gni (Kcal/mol) = 6.0507302 +/- 0.0560609 0.9545390
----- ENTER <CR> TO CONTINUE OR Q TO QUIT ----- :
Ani (Kcal/mol/[D]) = 3.0154808 +/- 0.0270514 0.9531400
Giu (Kcal/mol) = 3.9656738 +/- 0.0625010 0.9593446
Aiu (Kcal/mol/[D]) = 0.9886546 +/- 0.0149034 0.9578601
SLOPE OF NATIVE BASELINE = 0.1212495 +/- 0.0156092 0.8182932
Y-INT OF NATIVE BASELINE = 0.9931683 +/- 0.0110327 0.6779617
SLOPE OF UNFOLDED BASELIN = -0.0903052 +/- 0.0039520 0.9159318
Y-INT OF UNFOLDED BASELIN = 9.9357623 +/- 0.0298294 0.9304619
Z PARAMETER = 0.7530495 +/- 0.0035151 0.8453717
TEMPERATURE (K) = 298.1500000
>> BUFFER 2 [ 0.00000E+00] 3state.02.asc
Gni (Kcal/mol) = 6.0507302 +/- 0.0560609 0.9545390
Ani (Kcal/mol/[D]) = 3.0154808 +/- 0.0270514 0.9531400
----- ENTER <CR> TO CONTINUE OR Q TO QUIT ----- :
Giu (Kcal/mol) = 3.9656738 +/- 0.0625010 0.9593446
Aiu (Kcal/mol/[D]) = 0.9886546 +/- 0.0149034 0.9578601
SLOPE OF NATIVE BASELINE = 0.1361729 +/- 0.0201031 0.8488802
Y-INT OF NATIVE BASELINE = 0.9764976 +/- 0.0189499 0.6971704
SLOPE OF UNFOLDED BASELIN = -0.1036810 +/- 0.0089232 0.9258739
Y-INT OF UNFOLDED BASELIN = 10.0264725 +/- 0.0688860 0.9352576
Z PARAMETER = 0.2413364 +/- 0.0065508 0.8412052
TEMPERATURE (K) = 298.1500000
FILE FOR STATISTICS: 3state.fit.sta # This saves the output we just saw to a file. Values are correct.
SA> wp 3state.fit.par # save the parameter file. This is the most important thing to save.
SA> sca 1 2 1 # let's scan through the curves one by one (instead of overlaying).
SA> sd 2 # let's change the plot style so that residuals are drawn also.
SA> sca 1 2 1
SA> wr 3state.01.mod # we can save the fit curve and the residuals to separate xy ascii files.
WR> FORMAT (XY,XYE) [XY]:
WR> BUFFER# TO WRITE = [ 1 ]
WR> DATA(D), MODEL(M), OR RESIDUALS(R)?: m
SA> wr 3state.02.mod
WR> FORMAT (XY,XYE) [XY]:
WR> BUFFER# TO WRITE = [ 1 ] 2
WR> DATA(D), MODEL(M), OR RESIDUALS(R)?: m
SA>
SA> wr 3state.01.res
WR> FORMAT (XY,XYE) [XY]:
WR> BUFFER# TO WRITE = [ 1 ]
WR> DATA(D), MODEL(M), OR RESIDUALS(R)?: r
SA> wr 3state.02.res
WR> FORMAT (XY,XYE) [XY]:
WR> BUFFER# TO WRITE = [ 1 ] 2
WR> DATA(D), MODEL(M), OR RESIDUALS(R)?: r
SA>
SA>
SA> quit # exit Savuka and return to Unix shell.
