|
GeneMarker® MLPA
GeneMarker for Multiplex Ligation-dependent Probe Amplification
(MLPA™)
Download MLPA panels...click here.
MLPA was introduced by the Microbiology Research Center, Holland, in
January 2002 and has become a rapidly growing technique used in the detection
of exon deletions in BRCA1, MSH2 and MLH1 genes associated with breast
and colon cancer, as well as trisomies found in Down Syndrome 1,2,3. Copy
number changes in genomic DNA as well as mRNA profiling can be achieved
through this technique.
Developed in collaboration with the Pals laboratory of VU Medical Center,
the Netherlands, SoftGenetics developed an completely integrated application
for MLPA analysis, eliminating the time consuming data transfer of spreadsheet based analysis. GeneMarker, in the MLPA mode, converts data from
any sequencing system, offers two normalization and analysis methods,
and includes a customizable Patient Report.
GeneMarker now includes a Luminex-MLPA module that quickly and accurately analyzes data from Luminex instruments (Luminex 100 and Luminex 200). The software is compatible with the Luminex instrument data format (*.csv).
GeneMarker® Luminex-MLPA analysis automatically performs background subtraction, and flags suspect intensities according to user-specified thresholds. The software selects the least variable sample, from the sample set to act as the reference for determining copy number change within the patient sample set.
Panel Editor
The MLPA Panel Editor is designed to give users the ability to indicate
and label multiple internal controls and markers. Additionally the software
also features pre-made MLPA panels based on the MRC Holland probe sets.
These pre-made panels can be easily imported into the panel editor.
GeneMarker for MLPA analysis allows the selection
of multiple internal controls and markers. Panels based upon the
MRC kits are easily imported.
|
Data Normalization
Due to the variation of PCR efficiencies from small to large
DNA fragments or from sample to sample, two selectable normalization methods
are provided.
The first normalization method is the traditional method based upon the
normalization to the control probes.
The second method, unique to GeneMarker, normalizes peak intensities
based upon the statistically most probable median intensities. In order to
correct for the peak intensity variation over size, an exponential function
a*e-bz is used to fit to the square root of peak intensities, where z
is size, and a and b are fitting constants.
Normalization using the control probes is shown in figure 2. This correction
removes the trend of dropping intensities as the DNA fragment size increases,
and sets the height ratios of control probes to approximately 1. However,
the trend of peak intensities vary greatly from one sample to another
with the internal control probes (indicated as blue box). We have found
that the use of fewer control probes often results in large errors in
the intensity normalization.
 |
 |
Figure 2. Traditional Control Probe Normalization.
|
Figure 3. SoftGenetics’ Population Normalization.
|
SoftGenetics’ “Population Normalization” addresses
the above problems (Figure 3). Median peak intensities are derived from
the first nine data points, then sliding to data points 2-10, 3-11, etc.
to ascertain the local median intensities. Outliers are rejected after
applying a median filter. All of the probes (control and test) with the
median intensities are then used to fit the exponential function. This
methodology creates higher accuracy and lower false positive rates.
Data Analysis
Following normalization, the data is plotted in two formats: ratio and
regression type. First, the intensity ratios of identical probes from
the patient sample can be compared to that of the control samples. Deletions
and duplications appear as outliers from the data set, as points outside
of the threshold lines. (Figure 4)
The second regression plot method provides the peak intensity deviations
of the patient compared to that of the control sample, as defined by the
user. The square root of peak intensity is used to form the best fit of
the peak intensity deviation exponential function. To calculate the regression
and reject outliers: 80% of peaks with smaller deviations are retained
in the regression line calculation, and 20% are rejected initially. GeneMarker
then iterates multiple times to reject and retain peaks, with a confidence
of 95%. The 99% confidence level of the regression is shown within the
green lines; duplications > 1.33 and deletions < 0.75 of DNA copy
number are shown as red dots outside of the red lines. The points with
normal DNA copy number are green dots, and the control probes are blue
dots. (Figure 5)
 |
 |
Figure 4: The height ratio plot of a patient sample
compared to that of a control sample. |
Figure 5: The regression plot of the square root
of peak intensity deviation of a sample compared to a control. |
Patient Report
GeneMarker provides a patient report that can be printed or saved as a
PDF file.
Figure 6: The patient report includes sample ID,
analysis parameters, report, graph and electropherogram by sample.
Application Notes
MLPA application note [pdf]
|
|
Copyright © 2008