From LipidomicsWiki

Contents 1 2D-PAGE MS Proteomics 1.1 Input 1.2 Output 1.3 Processing 1.4 Proposals and discussion points 2 Protein chip 2.1 Input 2.2 Output 2.3 Processing 2.4 Proposals and discussion points 3 Implementation 4 Processing Pipeline

2D-PAGE MS Proteomics

Input

High resolution 2D PAGE Images including at minimum 6 biological and at minimum 2 technical replicates.

Output

Differential 2D Spots with regulation factor and significance value for further analysis and identification by mass spectrometry

Processing

For quantitative gel comparison the 2D gels were digitized and analysed using Proteomweaver 3.0 software. The determination of the total spot number was done with the image analysis software. Spot quantification and normalisation was performed with the software-implemented algorithms. Protein detection and spot matching were controlled visually. Protein spots with a saturated stain (so called donut spots) were selected manually and compared visually. Spots showing a standard deviation of more than 66 % within one group (gel triplicates), were excluded from the further analysis. If visual inspection of the spot revealed an experimental error based causation (e.g. local streaking), only the outlier was excluded, the two corresponding spots were still included in the set of analysed spots. If no experimental error based causation was found, the spot was completely excluded from further analysis. The total of well separated spots was finally selected for quantitative spot comparison. The spot intensities of the gel triplicates per sample were averaged. The mean spot intensities and CV for both patient groups were calculated and the mean ratio between Group A and B were determined. The quality of the gels allows to consider spots with an average spot intensity ratio >= 1.5 or <= 0.67 as potentially different between both groups. The statistical significane for all spots was checked by statistical tests.

First, statistical analyses were carried by a conventional Student’s t test with and without Bonferroni correction using MS Excel. For Bonferroni correction the significance level were recalculated for the pre-specified p values (p = 0.05, p = 0.01, and p = 0.001). Critical t values were calculated with and without Bonferroni correction for f = 18 degrees of freedom for N = 20 samples are given in Table 1.

For samples with less than 18 degrees of freedom the corresponding t values were
used [5].

Table 1: Critical t values used for Student’s t-test.
Critical t values
t (p = 0.05; f = 18) 2.101
t (p = 0.01; f = 18) 2.878
t (p = 0.001; f = 18) 3.922
tBonferroni (p = 0.000071; f = 18) 5.122
tBonferroni (p = 0.000014; f = 18) 5.886
tBonferroni (p = 0.0000014; f = 18) 7.045

Additionally, the data were evaluated by the software SAM (Significance analysis of
microarray data) [6]  on the basis of Student’s t-test and the non-parametric Wilcoxon [7]
signed-rank test.

5 Funk, W. et al. (1987) Statistische Methoden in der Wasseranalytik. VCH, Weinheim
6 Chu, G., Narasimhan, B., Tibshirani, R. & Tusher, V. (2002), Signifcance analysis of microarrays (sam)
software. Available: http://wwwstat.stanford.edu/~tibs/SAM/ via the Internet.
7 Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics, 1, 80-83

Proposals and discussion points

Write here any proposals and discussion points you may have.

Protein chip

Input

Clinically characterized serum or plasma samples obtained according to standard procedure, minimum requested amount 50 µl, within a study all samples should be of the same type, same sample system. Characterized samples are chosen with respect to the medical claim of interest. These samples are profiled on protein biochips such as UNIchip® AV-400 (Protagen AG, Dortmund, Germany).

Output

Relative units of (auto-)antibody binding of the antigens listed in the result file.

UNIchip protein biochip layout and principle of relative unit calculation:
(serial antigens dilutions in Field 3 are optional and not used within the scope of LipidomicNet)

Example of standard UNIchip report:

Processing

Protein biochips are manufactured by Protagen on nitrocellulose-coated slides. They are printed with a randomly selected set of 384 different recombinant human proteins representing different gene ontology classes as published (Feyen et al., 2008, Lueking et al. 2008). The proteins originate from an E. coli protein expression library and have been purified using a His-tag as published (Lueking et al., 2005). Each recombinant human protein is printed in quadruplicates on the biochip.
The protein biochips are incubated with serum or plasma samples as published by Lueking and co-workers (2005, 2006).
Read out of the results is performed with a confocal microarray reader (ScanArray 4000, Perkin Elmer Life Science) using identical settings for all biochips Image analysis is performed using the software package GenePix Pro 6.0 (Molecular Devices). Following image analysis the mean pixel intensity (median local background subtracted) is determined for each protein spot. From each biochip the median intensity of the 4 replica spots of each protein is determined. Based on all experiments for one sample, the average of the median intensities of each protein or protein concentration is reported. To compare the UNIchip® biochips the signal intensities obtained for the 400 human proteins are normalized using the IgG control spots for the secondary detection antibody as internal calibrator. As a novel method of binding profile calculation and standardization the average signal intensity of the IgG control spots for each antibody incubation is set to 100 %. The average normalized signal intensity of each of the human proteins is referred to this reference value leading to relative units in %.

Literature:
Angelika Lueking, Jens Beator, Elisabeth Patz, Stefan Müllner, Gabor Mehes, Peter Amersdorfer: Determination and Validation of Off-Target Activities of -CD44 Variant 6 Antibodies employing Protein Biochips and Tissue Microarrays. Biotechniques: 45 (4):P i-v (2008).

Oliver Feyen, Angelika Lueking, Axel Kowald, Christian Stephan, Helmut E. Meyer, Ulrich Göbel, and Tim Niehues: Off-target activity of TNF-α inhibitors characterized by Protein Biochips. Analytical and Bioanalytical Chemistry: in press (2008)/ online published 16.03.2008.

Sabine Horn, Angelika Lueking, Derek Murphy, Alexander Staudt, Claudia Gutjahr, Kirsten Schulte, Andrea König, Martin Landsberger, Hans Lehrach, Stephan B. Felix and Dolores J. Cahill: Profiling humoral auto-immune repertoire of dilated cardiomyopathy (DCM) patients and development of a disease-associated protein chip. Proteomics 6:605-613 (2006).

Angelika Lueking, Otmar Huber, Christopher Wirths, Kirsten Schulte, Karola M. Stieler, Ulrike Blume-Peytavi, Axel Kowald, Karin Hensel-Wiegel, Rudolf Tauber, Hans Lehrach, Helmut E. Meyer and Dolores J. Cahill: Profiling Alopecia areata autoantigens based on protein microarray technology. Molecular Cellular Proteomics 4: 1382-1390 (2005).

Claudia Gutjahr, Derek Murphy, Angelika Lueking, Andrea Koenig, Michal Janitz, John O’Brien, Bernhard Korn, Sabine Horn, Hans Lehrach, and Dolores J. Cahill: Mouse protein arrays from a TH1 cell cDNA library for antibody screening and serum profiling. Genomics 85:285-96 (2005).

Proposals and discussion points

The relative unit data of the standard UNIchip report can be further processed by:

1. determining the mean and the standard deviation of all relative units from a reference sample set
2. use the mean plus 2 (or 3) standard deviations as a cut-off value to determine statistically significant autoantibody levels in the disease sample set.

Implementation

Implementation of Proteomics SPPs

Processing Pipeline

Processing Pipeline for Proteomics SPPs

Back to Standard Processing Procedures