Feature Finder

Features:

• Calculates and removes the background from the spectra.

• Calculates the peak intensity, position, peak width, and area under the peak (positive and negative).

• Visualizes the peak in the dataset.

• Generates the new dataset for the peak of interest and saves it in the original directory.

• The toolbox supports parallel computing to accelerate computation.

Steps:

1.      Select the area of interest.

2.      Correct the spectrum for background using None, Asymmetric Least Squares Smoothing, Savitzky-Golay, or Bayesian EM methods

3.      Move the boundaries to select a spectrum and calculate spectral parameters: such as Peak Intensity, Peak Maximum, Peak Width, and Peak Area.

4.      Press the Find Pixels button to visualize the pixels that have the peak features. This can be done by two different methods specified below.

• Method 1: Find Peaks by Peak Intensity. Change the values of the peak to any positive value less or equal to the identified peak intensity. Enter zero, if you would like to see the pixels in the image corresponding to the entire peak. Press Apply to Data (Peak). 

• Method 2: Find Peaks by Peak Area. Leave the value intact. Press Find Pixels by Peak Area.

You can save the new datacube with the corresponding changes in the image by pressing Save the Datacube as a new dataset to the directory.

Additional Information:

Savitzky-Golay filter: The filter applies a type of sliding window to the data.

ORDER: is an order of polynomial.

WINDOW: is the number of data points in the window at any one time.

For example, Window = 11, and Order = 3 fits a polynomial of order 3 to 11 data points at a time and the ith data point will be approximated by the polynomial evaluated at the point corresponding to i.

Asymmetric Least Squares Smoothing: This method uses a smoother with an asymmetric weighting of deviations to get a baseline estimator. In doing such, this processor quickly ascertain and correct a baseline while retaining the signal peak information.

SMOOTHNESS: defines how smooth the baseline should be (default 100).

ASYMMETRY: defines how "low" the baseline should be. The range is from 0 to 1. Lower peaks require asymmetry approaching 0, while high peaks require asymmetry value approaching 1.

ITERATIONS: defines the number of iterations to reach the converge (default 20). The calculation time is proportional to the number of iterations.

Bayesian EM filter: The filter adjusts the variable background (baseline) of a signal with peaks by following three steps: 1) estimates the background within multiple shifted windows of width 200 separation units (s.u.) along the x-axis, 2) regresses the varying baseline to the window points using a spline approximation, and 3) adjusts the background of the input signal.

WINDOWSIZE: sets the width for the shifting window. The default is 200 s.u., which means a background point is estimated for windows of 200 s.u. in width.

STEPSIZE: sets the steps for the shifting window. The default is 200 s.u., which means a background point is estimated for windows at every 200 s.u.

REGRESSION METHOD sets the method used to regress the window estimated points to a soft curve. The default is 'pchip'; i.e., shape-preserving piecewise cubic interpolation. Other options are 'linear' and 'spline' interpolation.

ESTIMATION METHOD sets the method used to find the likely background value at every window. Default is 'quantile', in which the quantile value is set to 10. An alternative method is 'em', which assumes a doubly stochastic model.

SMOOTH METHOD sets the method used to smooth the curve of estimated points, useful to eliminate the effect of possible outliers. Options are 'none' (default), 'lowess' (Linear Fit), 'loess' (Quadratic Fit), or 'rlowess' and 'rloess' (Robust Linear and Quadratic Fit).

QUANTILE VALUE changes the default quantile value. The default is 0.10.

Note: The baseline subtraction mode does not preserve the height of the tallest peak in the signal when subtracting the baseline.

References:

Lucio Andrade and Elias Manolakos, "Signal Background Estimation and Baseline Correction Algorithms for Accurate DNA Sequencing" Journal of VLSI, special issue on Bioinformatics 35:3 pp 229-243 (2003)