3 Material and Methods


3.1 Material

3.1.1 Equipment




Centrifuge Biofuge stratos

Heraeus, Berlin, Germany

Centrifuge Fresco 17

Heraeus, Berlin, Germany

Confocal Microscope Fluo ViewTM 1000

Olympus, Hamburg, Germany

Fluorescence spectrometer (VC Eclipse)

Agilent Inc., Santa Clara, U.S.A.

Fridge (4 °C and -20 °C)

Bosch, Gerlingen, Germany

Fridge Hera freeze (-80 °C)

Thermo scientific, Berlin, Germany


Heraeus, Berlin, Germany

Microliter Cell

IMPLEN GmbH, Munic, Germany

Milli-Q Ultra Pure Water Purification System

Millipore Corp., Bellerica, U.S.A.

Laminar hood

Heraeus, Berlin, Germany

Pipette, electrical (pipetus)

Hirschmann Laborgeräte, Herrenberg, Germany

Photometer (Biophotometer plus)

Eppendorf, Hamburg, Germany

iQ5 Real-Time PCR Detection System

Bio-Rad Laboratories Inc., Hercules, U.S.A.


Thermo Scientific, Berlin, Germany

Vortex-genie 2

Scientific industries, New York, U.S.A.

Water bath DC10

Haake, Victoria, Australia

3.1.2 Consumable Material



Cryo tubes (1.5 ml)

Sarstedt, Nümbrecht, Germany

5 ml round-bottom tubes (FACS)

Sarstedt, Nümbrecht, Germany

Glass ware

Carl Roth, Karlsruhe, Germany


Eppendorf, Hamburg, Germany

Pipette tips (1 - 1000 μl)

Sarstedt, Nümbrecht, Germany

Reaction tubes (1.5 ml, 15 ml, 50 ml)

Sarstedt, Nümbrecht, Germany

Cell culture flasks (T25, T75)

Nunc, Langenselbold, Germany

Cell culture plates (6 well, 12 well)

Nunc, Langenselbold, Germany

35 mm glass bottom dishes

MatTek, Ashland, U.S.A.

Fluorescence quartz cuvettes (3x3 mm)

Hellma Analystics, Müllheim, Germany

3.1.3 Biological Material Eucaryotic Cell Lines

Cell line


BHK-21 cells (baby hamster kidney)

adherent, fibroblasts

MDCK cells (madin darbin canin kidney)

adherent, epithelial cells Viruses


Influenza A /Puerto Rico/34/8 and influenza A/X-31 (Institute of Biology, Humboldt University, Berlin) were grown in 10-day-old chicken embryos and purified as described previously [203].

Semliki Forest Virus and Vesicular Stomatitis Virus (kindly provided by PD Dr. M. Veit, Institute of Immunology and Molecular Biology, Free University, Berlin) were obtained in BHK-21 cells.

3.1.4 Chemicals 



Complete, Mini Protease Inhibitor Cocktail

Roche, Mannheim, Germany


Fluka Analytical, St. Louis, U.S.A.


Sigma Aldrich, St. Louis, U.S.A.


Th. Geyer, Renningen, Germany

Natrium chloride

Sigma Aldrich, St. Louis, U.S.A.

Nonidet P40 Substitute >99% (NP 40)

Fluka BioChemika, St. Louis, U.S.A.


Sigma Aldrich, St. Louis, U.S.A.

Propidium iodide

Sigma Aldrich, St. Louis, U.S.A.

RNasin Ribonuclease Inhibitor

Promega Corp., Madison, U.S.A.


Fluka, St. Louis, U.S.A.

Streptolysin O

Sigma Aldrich, St. Louis, U.S.A.

TRISMA Base >99.9% Tris[hydroxymethyl]aminomethane

Sigma Aldrich, St. Louis, U.S.A.

3.1.5 Media and Solutions




DMEM (without Phenol red)

PAA GmbH, Wien, Austria

DPBS (with Calcium and Magnesium)

PAA GmbH, Wien, Austria

DPBS (Without Calcium and Magnesium)

PAA GmbH, Wien, Austria

Foetal Bovine Serum (FBS)

PAA GmbH, Wien, Austria

HEPES (25 mM) in DPBS

PAA GmbH, Wien, Austria

L-Glutamat (200 mM)

PAA GmbH, Wien, Austria

Penicillin/Streptomycin (100x)

PAA GmbH, Wien, Austria

Trypsin/EDTA (0.025 %)

PAA GmbH, Wien, Austria

3.1.6 Kits



RNeasy Mini Kit (RNA purification)

Quiagen, Hilden, Germany

SuperScriptTM II (cDNA synthesis)

Invitrogen, San Diego, U.S.A.

3.1.7 Antibodies 



mouse anti-NP (Influenza A), monoclonal

Chemicon, Billerica, U.S.A.

goat anti-mouse IgG, conjugated with Alexa 568

Mol. Probes, Karlsruhe, Germany

rabbit anti-VSV glycoprotein, polyclonal, serum

PD Dr. M. Veit, Free University, Berlin, Germany

goat anti-rabbit IgG, conjugated with Cy3

Mol. Probes, Karlsruhe, Germany

3.1.8 FIT-PNAs and Molecular Beacon


FIT-PNAs were synthesized by Dr. A. Knoll (Institute of Chemistry, Humboldt University, Berlin, Germany) and the Molecular Beacon was received from BioTez GmbH (Berlin, Germany).





neuraminidase, H1N1



neuraminidase, H1N1



matrix protein, H1N1



L protein, VSV


MB 2

neuraminidase, H1N1


3.1.9 Oligomers and Primers

Nucleotides are indicated referring to the accession number NC_002018 (A/PR/8, H1N1), if not stated otherwise.




RNA target 3a

nt 599-615, A/PR/8, NA


RNA target 3b

nt 594-618, A/PR/8, NA


RNA target 3c

nt 525-539, A/PR/8, M1


RNA target 4

nt 16-32, H3N2, DQ874878


DNA target 5

nt 599-615, A/PR/8


RNA target 7

nt 9108-9124, VSV, 001560


primer fwd NA

nt 570-595


primer rev NA

nt 643-661


primer fwd M1

nt 461-482


primer rev M1

nt -595-615


3.1.10 Software and Web Pages



Adobe Photoshop 7.0

Adobe Systems Inc., San Jose, U.S.A.

Ape- A Plasmid Editor

M. Wayne Davis

Entrez PubMed



Tree Star, Ashland, U.S.A.

FluoView 1000

Olympus, Tokyo, Japan

GraphPad Prism

GraphPad Software, La Jolla, U.S.A.


Zuker et al. 1991 [204]

Microsoft Office 2007

Microsoft, Washington, U.S.A.


http://www.ncbi. nlm.nih.gov/blast/mmtrace.html

Gene Ontology Project


3.2 Methods

3.2.1 Working under Sterile Conditions


Preventing contamination by bacteria, yeast or fungi is one of the important aspects one has to consider while working with mammalian cell culture. Infections influence cellular processes and states and thus the behaviour within experimental treatments. All used glass and plastic ware was sterilised employing a steam autoclave. The laminar hood was cleaned by UV light for 1 h weekly. Prior to every working procedure the bench was treated with Meliseptol.

3.2.2 Cell Culture General Handling

MDCK and BHK-21 cells were obtained in DMEM (without phenol red) supplemented with 10% FBS, 1 mM L-glutamine and 1% penicillin /streptomycin in a humid incubator at 37 °C and 5% CO2.

For long-term storage cells were harvested using 2 ml trypsin/EDTA and transferred into freezing medium (DMEM without supplements, 20% DMSO, 80% FBS). To obtain a decrease in temperature by one degree per hour the cryo tubes have been incubated overnight at -80 °C using an isopropanole freezing box prior to storage in the liquid nitrogen tank.


Thawing was performed in a 37 °C pre-warmed water bath. Immediately after thawing the cells were transferred into 12 ml of fresh DMEM and centrifuged at 300 x g for 5 min. The cell pellet was resuspended in 5 ml pre-warmed DMEM and the cells were maintained in T25 cell culture flasks.

In a period of 3 to 4 days BHK and MDCK cells were splitted in a 1:10 ratio. Adherent growing cell lines establish connections with the plastic layer of the culture flasks since they are organised in association to their neighbouring cells in nature. Therefore a trypsin mediated peptide cleavage was necessary prior cell passaging.

Cells subjected to virus infection were seeded in 35 mm glass bottom dishes at 80% confluency in DMEM lacking phenol red. Usually, this indicator is used to assess the pH of the media. In fluorescence microscopy this red dye may disturb the fluorescence of interest or interfere in the emission range. Cell Viability Test


Due to virus infection or SLO mediated transient cell perturbation (see chapter the cell viability may be negatively influenced. Proofing cell viability cells were treated with 5 µg of propidium iodide in 1 ml DPBS (Ca2+ / Mg2+) for 5 min at room temperature and washed once with fresh DPBS (Ca2+ / Mg2+) prior imaging. Propidium iodide is a DNA intercalating fluorophore which is commonly used for identifying dead or apoptotic cells as it is excluded from viable cells. For a positive control 10 µl of 1 N HCl was added over the cells for 20 min resulting in cell apoptosis. Delivery of PNAs into Living Cells 

Caused by their structural characteristics (e.g. uncharged peptide backbone) PNAs do not interact with Lipofectamine 2000 (Invitrogen) or related transfection reagents. Therefore the delivery of PNAs into living MDCK cells requires the help of streptolysin O, a streptococcal haemolytic exotoxin, which mediates reversible plasma membrane penetration as described previously [121]. In brief, 5 U streptolysin O in DPBS supplemented with 25 mM Hepes and 10 mM DTT were activated for 90 min at 37 °C. For staining, 250 nM PNA in 500 µl activated SLO containing solution were added to the cells for 30 min at 37 °C. Resealing was performed by adding 2 ml of fresh DMEM and incubating for additional 30 min at 37 °C. Before starting fluorescence microscopy imaging the medium was changed to 1 ml DMEM without supplements. Delivery of PNAs into Fixed Cells 

MDCK cells are of epithelial origin and thus exhibit a strong cell-cell-adhesion. In contrast, the fibroblast cell line BHK-21 lacks such extensive adhesion and is therefore unsuitable for SLO treatment. For FIT-PNA delivery BHK-21 cells were fixed with the help of 1 ml 4% paraformaldehyde in DPBS. After 20 min incubation at room temperature, cells were rinsed with 1 ml DPBS before adding 1 ml 0.1% Triton X-100 for 10 min at room temperature. Two washing steps with 1 ml DPBS were required to stop further permeabilisation. PNAs were delivered at a concentration of 250 nM in DPBS to achieve an efficient staining of viral mRNA. In preparation of fluorescence microscopy imaging the staining solution was replaced with 1 ml fresh DPBS.


Fixed BHK cells in 35 mm cell culture dishes can be stored at 4 °C for several weeks with negligible decrease in fluorescence signal if protected from light. Delivery of PNAs into Fixed Cells in Solution 

For intracellular FACS staining cells have to be in suspension. Therefore adherent growing MDCK cells were treated with trypsin and for further staining sedimented at 300 x g for 3 min. All following steps were performed on ice to maintain 4 °C during the staining procedure. The cell pellet was resuspended in 1 ml 4% paraformaldehyde. Cells were fixed for 20 min, sedimented (300 x g, 3 min) and washed with 1 ml DPBS. Permeabilisation was performed in 500 µl 0.5% saponin in DPBS. The cells were centrifuged as described above to reduce the volume by decanting the supernatant. 250 pM of FIT-PNAs were added into the remaining cell solution (circa 50 µl) and incubated for 15 min. After staining, the cells were rinsed with DPBS and stored at 4 °C until FACS analysis. Cell Lysate 

In preparation of experiments in living cells FIT-PNAs were hold in cell lysate in order to assess their nuclease resistance. Cells seeded in 6-well plates were infected as indicated. After rinsing with 1 ml DPBS cells were lysed using 1 ml lysate buffer (50 mM Tris (pH 7.4), 150 mM NaCl, 0.8% NP-40 in dd H2O) for 5 min at room temperature. Carefully the cell lysate was transferred into 1.5 ml reaction tubes and stored on ice until fluorescence spectrometer measurement.


Data acquisition was performed in fluorescence cuvettes filled with 100 μl cell lysate from non-infected MDCK cells. Fluorescence was measured at 530 nm after excitation at 485 nm. For hybridisation 0.1 nmol FIT-PNA probe 1a and 1 nmol target 3a were incubated in water for 10 min. 5 μl of this solution as well as a control solution lacking the target were added to the lysate containing cuvettes. The fluorescence was measured at 37 °C. Data points were collected every minute for 30 min followed by a 5 min interval for additional 30 min.

For the Molecular Beacon MB2 the excitation wavelength was 559 nm and emission was recorded at 593 nm. The measurement was performed simultaneously to the described procedure of the FIT-PNA 1a.

The relative fluorescence (=responsiveness) was determined at various time points and calculated according to the following equation:


, where F and F 0   are the fluorescence intensity of target-containing or control lysates, respectively, and F (t=0) is the background fluorescence.

3.2.3 Infection Protocol

BHK-21 cells were infected with Vesicular Stomatitis Virus (subtype Indiana) and Semliki Forest Virus. MDCK cells were infected with influenza A/PR/8 virus. Cells were seeded 1 day prior to infection in 35 mm glass bottom dishes for microscopy imaging or in 6-well plates for e.g. FACS analysis at 80% confluence. In all indicated experiments infection with virus was performed at a multiplicity of infection (M.O.I.) of 100. To this end, the virus solution was added directly into 1 ml DMEM and added to the cells. To allow the virus to attach to the cell surface after one hour incubation at 37 °C, the supernatant was replaced by 1 ml fresh DMEM and incubation was continued.

3.2.4 Molecular Biology


To assess the viral mRNA production during the infection life cycle and to identify optimal time points for fluorescence microscopy imaging quantitative RT-PCR with influenza A specific primers was employed. RNA Purification 

Total RNA from influenza A/PR/8 (infection protocol see chapter 3.2.3) at various time points p.i. and non-infected MDCK cells was purified using the RNeasy Mini Kit following the manufactures instruction. In preparation for the purification 106 cells per sample were harvested by trypsinisation and sedimented at 300 x g for 5 min. Cell pellets were stored on ice until the experimental protocol was continued. RNA concentration was determined by an optical density (OD) measurement.  In vitro cDNA Synthesis 

As starting material for the in vitro reverse transcription 3 µg of total RNA was applied. The SuperScriptTM II and oligo(dT) primer were employed as recommended by the manufactures protocol. It is recommended to perform the cDNA synthesis immediately after the RNA extraction in order to prevent damage to the RNA by freezing and thawing processes. The synthesised cDNA was quantified by OD measurement. Real-Time Quantitative-PCR – Measurement


Measurement of the RT-qPCR was performed by Dr. Andrea Knoll (Institute of Chemistry, Humboldt University Berlin).

The reaction was carried out by using 100 ng cDNA of each sample as template employing 400 nM influenza A specific primers (see 3.1.9). This resulted in the amplification of a short sequence in the NA or M1 gene, respectively. FIT-PNA 1a or 1c was used at 500 nM (see 3.1.8) were utilized as detection fluorophore instead of the standard intercalator SYBR-Green enhancing the specificity of the data analysis. The reaction was realised in a volume of 20 µl in 96-well plates as described in table 1.

Table 1: RT-qPCR programme for cDNA amplification using the iQ5 Real-Time PCR Detection System.




initial denaturation

95 °C

3 min


95 °C

10 s

40 cycles


60 °C

20 s


72 °C

20 s


Fluorescence of the FIT-PNAs was measured during the annealing step every 3 s. As control cDNA was replaced by dd H2O. Real-Time Quantitative-PCR – Data Analysis

The calculation of specific viral mRNA copies per infected cell was based on the quantitative RT-PCR results. As mentioned above specific FIT-PNAs have been employed for signal detection.

The fluorescence intensity of TO or BO, respectively, correlates with the amplification of the template during the specific reaction in samples of infected MDCK cells. For quantification of specific target sequence (= copies of specific viral mRNA), a calibration curve was generated based on the related cycle of threshold (CT) values. The threshold fluorescence intensity was set threefold above the average no-template control fluorescence emission. The number of copies of specific viral mRNA present in 1 ng cDNA of starting material (100 ng cDNA for RT-qPCR) can be estimated according to the following equation:


, where N A is Avogrado`s number (6.022 x 1023 molecules /mole) and the molecular weight of the amplicon is 61652.4 g/mol.

Considering the total amount of synthesised cDNA the copy number of specific viral sequence per total cDNA was calculated. With the help of the total amount of purified RNA with regard to the starting material for the in vitro reverse transcription, the copy number per total RNA was determined. Assuming that 106 cells have been used one can roughly estimate the specific viral mRNA concentration per infected cell.

3.2.5 Immunocytochemistry (ICC)


Infected MDCK cells were stained 18 h p.i. with an anti-nucleoprotein (NP) antibody of influenza A virus to verify infection efficiency. For this purpose the intracellular labelling protocol using saponin as detergent was adapted. Cells were fixed and permeabilised as described (see chapter but stayed attached in the cell culture dish. The fixed and permeabilised cells were rinsed twice with 1 ml DPBS for 1 min before blocking with 1 ml 3% BSA in DPBS for 1 h at 4 °C. The anti-NP antibody was diluted 1:1000 in 3% BSA in DPBS. There was no washing step required prior to incubation with the anti-NP antibody. The blocking buffer was replaced by 100 µl of the antibody solution and incubated for 1 h at room temperature. For visualisation of the antibody binding and thus indirect NP presence verification a secondary anti-mouse IgG antibody conjugated with Alexa 568 was added over the cells after rinsing the cells twice with 3% BSA in DPBS. For an efficient staining, 100 µl of the secondary antibody diluted 1:1000 in 3% BSA in DPBS was sufficient. After 1 h incubation at room temperature cells were washed twice with DPBS without supplements to prevent any interference or detaining background signals during microscopy imaging.

Verification of an efficient VSV infection in BHK-21 cells was performed following the same protocol. Specifically, the rabbit anti-VSV glycoprotein antibody was diluted 1:100 in 3% BSA in DPBS. The ICC was started 3 h p.i. to visualize VSV glycoprotein expression.

3.2.6 Fluorescence Activated Cell Sorting (FACS)

Cells can be categorised and identified based on cellular characteristics such as gene expression or status of differentiation (cluster of differentiation) using extracellular FACS analysis. FACS is a standard tool in clinical diagnostics for several applications. The applicability of the FIT-PNA technique to FACS analysis was investigated. For this purpose, influenza A infection was determined employing viral mRNA specific FIT-PNAs which were delivered into fixed MDCK cells as described in chapter


The cell solution was transferred into FACS tubes and analysed using a BD FACS AriaTM II. TO fluorescence was recorded using the FITC channel with an excitation of 494 nm and an emission maximum of 519 nm. Each measurement was performed with 10,000 counting events. The system was calibrated with non-infected and unstained control MDCK cells to define the gate of intact cells in regard to the FCS (forward scatter, cell size) and SSC (sideward scatter, granular appearance) values. The settings were constant. FACS data were analysed using the provided software FlowJo.

3.2.7 Confocal Laser Scanning Microscopy (CLSM)

Fluorescence image acquisition was performed by a FluoView TM 1000 scanning unit using an IX 81 inverted confocal microscope and a 60x oil-immersion objective or a 60x water-immersion objective as indicated. The resolution was set to 512 x 512 pixel and sampling speed to 40 µs per pixel. Fixed cells were imaged at room temperature while for living cells a climate chamber to maintain 37 °C was required. Excitation and emission was performed as listed in table 2.

Table 2: Excitation and emission range of fluorophores used in FIT-PNAs or immunocytochemistry applications.



Emission range


448 nm

510 nm - 540 nm


440 nm

470 nm - 500 nm

TO and BO (sequential)

TO: 448 nm, BO: 440 nm

TO: 530 nm - 600 nm, BO: 460 nm - 490 nm

Alexa 568

559 nm

570 - 670 nm


559 nm

565 - 655 nm

3.2.8 Stable Isotope Labelling of Amino Acids in Cell Culture (SILAC) – Measurement


The following experimental part of the SILAC approach was realized in collaboration with Dr. Björn Schwanhäußer (MDC, Berlin) who kindly provided the SILAC-media and performed the liquid chromatography mass spectrometry (LC-MS) including required material preparation. 

In global proteomic analysis on a systems level, recently SILAC attracted attention as a widely applicable and practical technique in combination with LC-MS.

The strategy of a typical SILAC experiment is premised on the metabolic incorporation of stable isotopic variants of amino acids. On this account, the applied FBS had to be dialysed to eliminate the natural amino acids which shall be replaced by their labelled counterpart. Here, a medium heavy and a heavy SILAC cell culture medium were prepared containing 84 mg/l 13C6-L-argine in combination with 146 mg/l N4-L-lysine or 84 mg/l 13C615N4-L-arginine in combination with 146 mg/l 13C615N4-L-lysine, respectively. The light version of SILAC medium was supplemented with the non-labelled amino acids.


Upon cultivation in the specific SILAC medium over a period of eight passages MDCK cells incorporated uniformly the isotopic amino acids in their proteome. There was no influence on growth rate or cell viability provoked by the treatment of MDCK cells with the SILAC media.

MDCK cells were seeded in 6 well plates, infected as described (see chapter 3.2.3) with influenza A/PR/8 and harvested by trypsinisation according table 3 creating a time range 0 to 8 h p.i.

Table 3: Scheme of the experimental procedure of influenza A/PR/8 infected MDCK cell preparation within the SILAC approach.


Harvesting time point

MDCK cells in light SILAC medium

0 h post infection

MDCK cells in medium heavy SILAC medium

1 h, 2 h post infection

MDCK cells in heavy SILAC medium

4 h, 8 h post infection


After centrifugation at 300 x g for 3 min cell pellets were stored at -20 °C and mixed according to the following scheme:

Sample I:

0 h p.i. (light) + 1 h p.i. (medium heavy) + 4 h p.i. (heavy)

Sample II:

0 h p.i. (light) + 2 h p.i. (medium heavy) + 8 h p.i. (heavy)

After urea/thiourea mediated cell lyses an in-solution digest with trypsin and lysine-C protease which preferentially cleaves after lysine and/or arginine residues was performed. Peptides were desalted and concentrated with the help of self-prepared C18-columns. These elution columns operate on the principle of hydrophobic forces resulting from the binding of the peptide molecule to the C18-chain (stationary phase) upon association with the ligand in the aqueous eluent (mobile phase). Reduction of the water surface tension by adding less polar solvents (like methanol) leads to a decrease in retention of the bound peptide molecule.


Accordingly, the peptides are separated by their hydrophobic character using high performance liquid chromatography (HPLC) which automatically reduces the polarity of the mobile phase during the course of analysis producing a length gradient elution of the peptides. Immediately, the solute peptides were electrostatically ionized by electrospray ionization (ESI) and analysed by applied electrical fields inside the mass spectrometer. The peptide sequence was assessed by data analysis obtained from peptide collision with gas molecules.

3.2.9 Stable Isotope Labelling of Amino Acids in Cell Culture (SILAC) – Data Analysis

Data analysis of the LC-MS measurement was realized in collaboration with Dr. Björn Schwanhäußer (MDC, Berlin) and Max Flöttmann (Humboldt University, Berlin).

Light, medium heavy and heavy variants of a peptide were represented as clusters in an MS-output plot and discriminated due to their mass difference generated by the isotopic atoms. Quantification analysis was performed using the Max-Quant software [205] based on the formation of ratios within each isotope cluster. The measured peptides were aligned with dog protein databases (Mascot, Uniprot) which were in silico digested with the aforementioned proteases. Identified peptides were assigned to their biological process using gene ontology (GO) enrichment analysis via a hypergeometric test [206] package in Bioconductor for individual clusters with the complete set of measured genes as a background distribution.


The Gene Ontology Consortium provides an appropriate source of annotations for gene products and their properties. These include a cellular component, the molecular function and the biological process relevant in living systems like cells, tissues or organs. To exemplary elucidate this annotation the GO terms of cytochrom c are the following: mitochondrial matrix and mitochondrial inner membrane (= cellular component terms), oxidoreductase activity (=molecular function term) and oxidative phosphorylation and induction of cell death (=biological process terms).

The requirement of a hypergeometric test results from the structure of the given data set: the MS analysis offers a peptide list which is connected to the corresponding GO term list.

The amount of genes in a given list (cluster) determines the value of expected frequency how often a certain GO term was assigned to this cluster. The probability that the expected value matches with the real one is expressed as p-values ranging from 0 to 1. In the hypergeometric test a GO term receives a low p-value if the real value exceeds the expected one.


Proteomic phenotyping for GO terms of the time point 8 h p.i. was performed as described in Pan et al. 2009 [194]. The distribution of measured log2 fold changes was divided into 4 quantiles. For each quantile a GO term enrichment test was repeated with all measured genes as background.

If a GO term reached a p-value > 0.5 for one of the quantiles it was transformed by -log10(p-value) and standardized using the following equation:


, where z is the standard score (dimensionless), x is the raw score (= p-value), mean(x) is the mean of all GO terms, sd (x) is the standard deviation.

The result of this conversion process were basically used to compute a 2D display of all values of the given data matrix. In a so called heat map the z-transformed p-values are hierarchically presented in a colour-coded diagram.

Clustering of GO terms was realized by standardizing the log2 fold-changes for the whole recorded infection time course of 8 h running a soft clustering algorithm, the fuzzy-c-means from the Mfuzz package [207] in Bioconductor [208]. This method provides on the one hand a very noise robust performance and on the other hand simplification of further filtering of relevant genes for each cluster. Accordingly, the membership value (0 – 1) is designated for each protein to the corresponding cluster and also subject to the fuzziness value c each appoint can be assigned to more than one cluster. The membership value defines the distance of a gene from the cluster centre. The number of clusters (c = 6) was defined by an iterative approach.

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML-Version erstellt am: