Limits...
Novel approaches to smoothing and comparing SELDI TOF spectra.

Meleth S, Eltoum IE, Zhu L, Oelschlager D, Piyathilake C, Chhieng D, Grizzle WE - Cancer Inform (2005)

Bottom Line: We also calculated the 'Area under the Curve' (AUC) spanned by each window.Keeping everything else constant, such as pre-processing of the data and the classifier used, the AUC performed much better as a metric of comparison than the peaks in two out of three data sets.In the third data set both metrics performed equivalently.

View Article: PubMed Central - PubMed

Affiliation: Department of Medicine, Medical Statistics Section, University of Alabama at Birmingham,Birmingham, Alabama, USA. Sreelatha.Meleth@ccc.uab.edu

ABSTRACT

Background: Most published literature using SELDI-TOF has used traditional techniques in Spectral Analysis such as Fourier transforms and wavelets for denoising. Most of these publications also compare spectra using their most prominent feature, i.e, peaks or local maximums.

Methods: The maximum intensity value within each window of differentiable m/z values was used to represent the intensity level in that window. We also calculated the 'Area under the Curve' (AUC) spanned by each window.

Results: Keeping everything else constant, such as pre-processing of the data and the classifier used, the AUC performed much better as a metric of comparison than the peaks in two out of three data sets. In the third data set both metrics performed equivalently.

Conclusions: This study shows that the feature used to compare spectra can have an impact on the results of a study attempting to identify biomarkers using SELDI TOF data.

No MeSH data available.


Normal AUCs 7000—10000 M/Z.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC2657649&req=5

f2a-cin-01-78: Normal AUCs 7000—10000 M/Z.

Mentions: The first data set consisted of SELDI spectra obtained from the sera of 21 women with normal pap smears and of 21 women with HSIL. Figure 1a shows the averaged spectra in the Normal and HSIL. The spectra are fairly similar. In Figure 1b is a closer view of the spectra between 7200 m/z and 9600 m/z. Figure 1c displays the same region (7200m/z to 9600 m/z) of the original spectrum, and the aligned, smoothed spectrum. This figure suggests that the alignment and feature selection using maximums, works to reduce the dimensionality of the data to one tenth of the original data (10458 m/z values to 1074), but keeps the main characteristics of the original spectrum. Figure 2a represents the trace of the maximum intensities between 7000m/z and 10000m/z region of the spectrum. Figure 2b is the trace of the AUCs between 7000m/z and 10,000m/z. As one can see, the AUC trace also provided a fairly good summary of the original spectrum, although the magnitude of the AUCs are as expected much larger than the values of just the maximums at a particular m/z value. These findings were repeated in the other two data sets used in the study.


Novel approaches to smoothing and comparing SELDI TOF spectra.

Meleth S, Eltoum IE, Zhu L, Oelschlager D, Piyathilake C, Chhieng D, Grizzle WE - Cancer Inform (2005)

Normal AUCs 7000—10000 M/Z.
© Copyright Policy - open-access
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC2657649&req=5

f2a-cin-01-78: Normal AUCs 7000—10000 M/Z.
Mentions: The first data set consisted of SELDI spectra obtained from the sera of 21 women with normal pap smears and of 21 women with HSIL. Figure 1a shows the averaged spectra in the Normal and HSIL. The spectra are fairly similar. In Figure 1b is a closer view of the spectra between 7200 m/z and 9600 m/z. Figure 1c displays the same region (7200m/z to 9600 m/z) of the original spectrum, and the aligned, smoothed spectrum. This figure suggests that the alignment and feature selection using maximums, works to reduce the dimensionality of the data to one tenth of the original data (10458 m/z values to 1074), but keeps the main characteristics of the original spectrum. Figure 2a represents the trace of the maximum intensities between 7000m/z and 10000m/z region of the spectrum. Figure 2b is the trace of the AUCs between 7000m/z and 10,000m/z. As one can see, the AUC trace also provided a fairly good summary of the original spectrum, although the magnitude of the AUCs are as expected much larger than the values of just the maximums at a particular m/z value. These findings were repeated in the other two data sets used in the study.

Bottom Line: We also calculated the 'Area under the Curve' (AUC) spanned by each window.Keeping everything else constant, such as pre-processing of the data and the classifier used, the AUC performed much better as a metric of comparison than the peaks in two out of three data sets.In the third data set both metrics performed equivalently.

View Article: PubMed Central - PubMed

Affiliation: Department of Medicine, Medical Statistics Section, University of Alabama at Birmingham,Birmingham, Alabama, USA. Sreelatha.Meleth@ccc.uab.edu

ABSTRACT

Background: Most published literature using SELDI-TOF has used traditional techniques in Spectral Analysis such as Fourier transforms and wavelets for denoising. Most of these publications also compare spectra using their most prominent feature, i.e, peaks or local maximums.

Methods: The maximum intensity value within each window of differentiable m/z values was used to represent the intensity level in that window. We also calculated the 'Area under the Curve' (AUC) spanned by each window.

Results: Keeping everything else constant, such as pre-processing of the data and the classifier used, the AUC performed much better as a metric of comparison than the peaks in two out of three data sets. In the third data set both metrics performed equivalently.

Conclusions: This study shows that the feature used to compare spectra can have an impact on the results of a study attempting to identify biomarkers using SELDI TOF data.

No MeSH data available.