Version: 4.

11
Automated

peak separation PeakFit is the automatic choice for spectroscopy, chromatography or electrophoresis
PeakFit is the automatic choice for spectroscopy, chromatography or electrophoresis
and analysis
FOR SPECTROSCOPY FOR CHROMATOGRAPHY
PeakFit lets you accurately detect, separate and quantify hidden peaks PeakFit includes 8 different built-in
that standard instrumentation misses. PeakFit includes 18 different equations for asymmetric peaks typically
nonlinear spectral application line shapes, including the Gaussian, the found in chromatography data: the
Lorentzian and the Voigt - even a Exponentially-Modified Gaussian, the
Gaussian plus Compton Edge model for Haarhoff-Van der Linde, the NLC, the
fitting Gamma Ray peaks. As a product Giddings and even a Half-Gaussian
of the curve fitting process, PeakFit Modified Gaussian (an experimental
reports amplitude (intensity), area, center model similar to the EMG, but used to
and width data for each peak. Overall describe intra-column originated
area is determined by integrating the asymmetry). The peak area is computed directly as a parameter within
peak equations in the entire model. each of these functions, ensuring accuracy and enabling computation of
PeakFit can even deconvolve your confidence limits. PeakFit reports column efficiency, resolution, first
spectral instrument response so that you moment (center of mass), second moment, center (mode), peak width at
can analyze your data without the smearing that your instrument base and half maximum, and asymmetry at base and 10% of maximum.
introduces. PeakFit can even deconvolve your chromatographic detector response,
so you can analyze your data without instrument induced asymmetry.
PeakFit also supports AIA data file input.
FOR ELECTROPHORESIS
PeakFit gives you the ability to
quickly and easily separate, locate FOR SIGNAL COMPONENT ANALYSIS

www.systat.com
and measure up to 100 peaks Even though PeakFit is widely used
(bands), even if they overlap. With by chemists, biologists, etc., many
82 nonlinear peak models to electrical engineers have found
choose from, you're almost PeakFit's statistical nonlinear fitting
guaranteed to find the best equation techniques extremely useful for
for your data. The sophisticated separating overlapping signals.
array of baseline types lets you "PeakFit has been a big time saver. I wish
integrate only the significant portion of the bands in your data. The I had found this product years ago"
results of these measurements are then automatically recorded in a - John Spradling,
PeakFit- generated report, or they can be shown graphically. PeakFit Communications Engineer
with a hand scanner can often replace $10,000 electrophoresis
instruments with better results!
TM

PeakFit
Automated
PeakFit's state-of-art-nonlinear curve fitting is essential for accurate peak analysis and conclusive findings
PeakFit's state-of-art-nonlinear curve fitting is essential for accurate peak analysis and conclusive findings
peak separation

and analysis WHY SHOULD YOU USE NONLINEAR CURVE FITTING? HIGHLY ADVANCED BASELINE SUBTRACTION
Nonlinear curve fitting is by far the most accurate way to reduce noise In this example, PeakFit's non-parametric baseline fitting routine easily
and quantify peaks. Many instruments come with software that only removes the complex background of a DNA electrophoresis sample.
approximates the fitting process by simply integrating the raw data PeakFit can also subtract 8 other built-in baseline equations, or it can
numerically. When there are shouldered or hidden peaks, a lot of noise, subtract any baseline you've developed and stored in a file.
or a significant background signal, this can lead to the wrong results. For
example, a spectroscopy data set may appear to have a peak with a FULL GRAPHICAL
"raw" amplitude of 4,000 units - but may have a shoulder peak that PLACEMENT OF PEAKS
distorts the amplitude by 1,500 units. This would be a significant error.
If PeakFit's auto-placement features
PeakFit helps you separate overlapping peaks by statistically fitting
fail on extremely complicated or noisy
numerous peak functions to one data set, which can help you find even
data, you can place and fit peaks
the most obscure patterns in your data. The background can be fit as a
graphically with only a few mouse
separate polynomial, exponential, logarithmic, hyperbolic or power
clicks. Each placed function has
model. This fitted baseline is then subtracted before peak
"anchors" that adjust even the most
characterization data (such as areas) are calculated, which gives much
highly complex functions,
more accurate results. And, any noise (from electrophoretic gels or
automatically changing that function's
Raman spectra) that might bias raw data calculations is filtered simply
specific numeric parameters. PeakFit's
by the nonlinear curve fitting process. Nonlinear curve fitting is
graphical placement options handle even the most complex peaks as
essential for accurate peak analysis and accurate research.
smoothly as Gaussians.
PEAKFIT OFFERS SOPHISTICATED DATA
PUBLICATION-QUALITY GRAPHS AND DATA OUTPUT
MANIPULATION

www.systat.com
Every publication-quality graph on this page was created using PeakFit's
With PeakFit's visual FFT filter, you can inspect your data stream in the builtin graphics engine - which now includes print preview and
Fourier domain and zero higher frequency points - and see your results extensive file and clipboard export options. The numerical output is
immediately in the time-domain. This smoothing technique allows for customizable so that you see only the content you want.
superb noise reduction while maintaining the integrity of the original
data stream. PeakFit also includes an automated FFT method as well as
PEAKFIT SAVES YOU PRECIOUS RESEARCH TIME
Gaussian convolution, the Savitzky-Golay method and the Loess
algorithm for smoothing. AI Experts throughout the smoothing options For most data sets, PeakFit does all the work for you. What once took
and other parts of the program automatically help you set many hours now takes minutes - with only a few clicks of the mouse! It's so
adjustments. And, PeakFit even has a digital data enhancer, which helps easy that novices can learn how to use PeakFit in no time. And if you
analyze sparse data. Only PeakFit offers so many different methods of have extremely complex or noisy data sets, the sophistication and depth
data manipulation. of PeakFit's data manipulation techniques is unequaled.

TM

PeakFit
Automated
A TRUE VOIGT FUNCTION... NONLINEAR PEAK SEPARATION AND ANALYSIS FOR
In Spectroscopy, instrument and Doppler broadening effects create a SPECTROSCOPE, CHROMATOGRAPHY AND
peak separation Gaussian line shape, while natural and collision broadening cause a ELECTROPHORESIS
Lorentzian line shape. The Voigt function is a convolution of both the PeakFit separates and analyzes nonlinear peak data better, more
Gaussian and Lorentzian functions. Most analysis packages that offer a accurately and more conveniently than your lab instrument
and analysis Voigt function use an approximation with very limited precision. here's why:
PeakFit actually uses a closed-form solution to precisely calculate the
Advanced nonlinear curve fitting
function analytically. PeakFit has four different Voigt functions, so you
techniques separate overlapping
can fit the parameters you're most interested in, including the individual
peaks that standard instrumentation
widths of both the Gaussian and Lorentzian components, and the
misses.
amplitude area of the Voigt
function. PeakFit's precise Peak finding and fitting is fully
calculation of the Voigt automated! PeakFit has three
function is crucial to the different procedures to accurately
accuracy of your analysis. characterize normal and hidden
peaks.
This pure voigt was created Graphically fit peaks yourself when you want to override PeakFit's
using PeakFit's data
transform feature
automation.
Highly advanced data smoothing and manipulation techniques
accurately process very noisy data.
PeakFit's automation makes it as easy as 1-2-3 Revolutionary constant zero second derivatives base-line fitting routine
PEAKFIT AUTOMATICALLY PLACES PEAKS IN THREE finds the best baseline. PeakFit also includes a non-parametric model for
WAYS extremely complex backgrounds.
PeakFit uses three procedures to automatically place hidden peaks; 82 built-in nonlinear equations (18 for spectroscopy and 8 for
while each is a strong solution, one method may work better with some chromatography) and up to 15 user-defined functions fit virtually all

www.systat.com
data sets than the others. data.
1. The Residuals procedure initially places peaks by finding local Fit up to 100 peaks at a time.
maxima in a smoothed data stream. Hidden peaks are then optionally 32-bit power and new Windows 95 interface plus, PeakFit is fully
added where peaks in the residuals occur. capable on Windows 3.1 and NT. With PeakFit, your accuracy will
2. The Second Derivative procedure searches for local minima within a increase, and you'll find and fit your peaks faster than ever. Don't guess
smoothed second derivative data stream. These local minima often or rely on the accuracy of non-dedicated software made by hardware
reveal hidden peaks. manufacturers. PeakFit far surpasses any other peak analysis package
3. The Deconvolution procedure uses a Gaussian response function with made!
a Fourier deconvolution/filtering algorithm. A successfully deconvolved
spectrum will consist of "sharpened" peaks of equivalent area. The goal "We know of no other comparable program that offers such sophisticated
is to enhance the hidden peaks so that each represents a local maximum. data analysis..."
Journal of the American Chemical Society Vol. 114, No. 20

TM

PeakFit
 PeakFit Features
Automated Data input 83 built-in nonlinear functions Numerical review
ASCII Spectroscopy (18): Gauss Amp, Gauss area, Lorentz Peak characterization data: center, amplitude, integrated
Excel amp, Lorentz area, Voigt amp, Voigt area, Voigt amp area, analytical area, FWHM, FW10, FWBASE,
Lotus 123 approx, Voigt amp G/L, Voigt area G/l, Gauss Cnstr asymmetry at HM, asymmetry at 10%, first and second
peak separation amp. Gauss Cnstr area, Pearson VII amp, Pearson VII moments, column efficiency and resolution, percentage
Quattro Pro Windows
SigmaPlot area, Gauss+Lor area, Gauss*Lor, Gamma Ray, areas, overlap areas
AIA chromatography Compton Edge Parameter statistics: parameter values, confidence limits
and analysis Chromatography (8): HVL, NLC, Giddings, EMG, (90,95,99%), t-values, standard errors
dBase III+, IV Fit statistics: Analysis of Variance, F-statistic, overall
DIF GMG, EMG+GMG, GEMG,GEMG5-parm
Statistical (31): Log Normal Amp, Log Normal Area, standard error, r2 value
ASCII and spreadsheet-like editors
Averaging digital filter Logistic Amp, Logistic area, Laplace amp, Laplace Data statistics: residual values, predicted y-values,
area, extr value amp, extr value area, log normal 4 amp, confidence/prediction intervals
Data preparation log normal 4 area, eval4 amp ailed, eval4 area tailed, Output and expert options
Gaussian deconvolution to remove spectrophotometer eval4 amp frtd, eval4 area frtd, gamma amp, gamma File export with full Generated Data: Lotus 123, Excel,
instrument response smearing area, inv gamma amp, inv gamma area, Weibull amp, QuattroPro Windows, SigmaPlot and ASCII
Weibull area, error amp, error area, chi-sq amp, chi-sq Graphs to clipboard or file in BMP or WMF formats
Exponential deconvolution to remove chromatographic
area, student t amp, student t area, beta amp, beta area,
detector response smearing
F variance amp, F variance area, Pearson IV All numeric data in graph to clipboard in spreadsheet
Smoothing (Savitsky Golay, FFT Filter, Loess, format
Gaussian convolution) General peak (12): erfc pk, pulse pk, LDR pk, asym
Real-time FFT/time domain graphical editor lgstc pk, lgstc pow pk, pulse pow pk, pulse wid2 pk, The Essential Desktop Tools for Scientists and Engineers
intermediate pk, sym dbl sigmoid, sym dbl Gausscum,
Dual-graph data sectioning with graphical data point SYSTAT , More graphs, more statistics, less effort
exclusion asymdbl sigmoid. AutoSignal, Transform time in no time
Transition (14): sigmoid asc, sigmoid desc, Gausscum
Non-parametric digital filter to filter or augment data TableCurve 2D , Automated curve fitting and equation discovery
asc, Gausscum desc, Lorentzcum asc, Lorentzcum desc,
Compare with reference TableCurve 3D , Automated surface fitting and equation discovery
lgstcdose rsp asc, lgstcdose rsp desc, lognormcum asc,
Subtract baseline imported from file PeakFit , Automated peak separation and analysis
lognorrncum desc, extrcalcum asc, extrcalcum desc,
Data transforms
pulsecum asc, pulsecum desc
Area normalization
Systat Software Inc.

Americas
Inspect second and fourth derivatives User defined functions
Data weighting Up to 10 parameters per function 501, Canal Boulevard,
Up to 15 UDF's active during fit Suite F Richmond,
Estimates can contain formulas and constraints CA 94804 - 2028 USA
Peak autoplacement Phone: 1 - 866 -797 - 8288
Automatic by local maxima and residuals Extensive mathematical, statistical, and logic functions
Fax: 1 - 510 - 231 - 4789
Automatic by second derivatives Baseline fit and subtract Fax:+1 - 510 - 412 - 2899
Automatic by deconvolved local maxima Automatic detection of baseline points by constant Email: [email protected]
Graphical placement and adjustments Systat Software UK Ltd.
second derivatives

Europe
Manual parameter adjustments Real-time fitting in conjunction with data point 23, Vista Centre, 50, Salisbury Rd
Share and lock parameters selection, deselection Hounslow - TW4 6JQ
Constant or variable widths and or shape in a single step Background functions (10): constant, linear, progressive LONDON, UK
Phone: +44-(0)208-538 0128
Non-linear curve fitting linear, quadratic, cubic, logarithmic, exponential, Fax: + 44-(0)208-538 0273
Marquardt-Levenburg algorithm power, hyperbolic and nonparametric
Email: [email protected]

Asia Pacific
Least-squares and 3 robust (maximum likelihood)
Graphical review Systat Software Asia Pacific Pvt. Ltd.
methods Component and sum curve graphs 5th floor, C-wing, Golden Enclave,
Up to 100 peaks and 1000 parameters Residuals graphs, including distribution and stabilized Airport road, Bangalore - 560017 INDIA
Intelligent constraints to insure fit integrity normal probability plots Phone: +91-080-5201604/5/6
Sparse curvature matrix for faster fitting
TM Confidence and prediction intervals Fax: +91-080-5201626
Both numeric and graphical fitting options Peak labels (amplitude, center, or area) Email: [email protected]
Zoom-in or toggle points during fitting
PeakFit System Requirements Microsoft Windows*95,98 and NT Pentiums or clone and above; 32MB RAM minimum (64MB RAM for wavelet and production facility recommended); 25MB hard disk space; SVGA or better