Proteinquantifizierung Standardisierung

Bioinformatik

Proteinquantifizierung Standardisierung

Bioinformatik

W. Huber, Div. Molecular Genome Analysis, DKFZ

Proteinquantifizierung:

U. Korf, Deutsches Krebsforschungszentrum (DKFZ)

T. Nann, Freiburger Materialforschungszentrum (FMF)

Bioinformatik:

W. Huber, DKFZ

Proteinquantifizierung:

U. Korf, Deutsches Krebsforschungszentrum (DKFZ)

T. Nann, Freiburger Materialforschungszentrum (FMF)

Bioinformatik:

W. Huber, DKFZ

mathematical modeling

Standardized procedures: experiments

data management modeling

biological system

quantification technology

statistical analysis

spin offs

This talk

W. Huber, Div. Molecular Genome Analysis, DKFZ

Protein quantificationthe parallelized Western Blot

Protein labelingaccuracy in detection

Bioinformaticsstandardization, statistical analysisdata exchange

Method Sample Pro Contra

Western-Blot

protein Well-establishedSimple protocol

Time-consumingSmall number of samplesQuantification is hard to standardize

2D-Gel 

complex protein mixture 

High resolution Time-consumingLimited pI/MW-range onlyIdentification: mass spectrometry

Mass Spec (ICAT)

protein no need for antibodiesrelative concentration

Expensive- hardware - personnel

ELISA protein measure protein activity and interaction 

high sample consumptionslow

Antibody Array

10 proteins relative or absolute (with internal standards)detection of PTMs(phophorylation)

method in development phase

Protein quantification methods

Economist Mar 15, 2003

„Studying proteins has long be a slow, arduous process.

Protein chips ... [promise to do for biology] ... what microprocessors did for personal computing.“

Sample Preparation

Data Analysis

Antibodies

RecombinantProteins

Spotting (384 well-plate)(25 Arrays/Run)

Immobilizationof Proteins

BlockingProcedure

Assay Scanning

Protein Array Technology

1. Fluorescent labeling (2 colors)

2. Relative quantification: abundance ratios lysate 1 / lysate 2

3. Absolute quantification: lysate / dilution series of known reference

4. Quantification of protein phoshporylation: phosphospecific antibodies

Protein quantification

2 5 10 20 50 100

500

2000

1000

0

subarray 1

PKA ng/ml

fluor

esce

nce

2 5 10 20 50 100

500

2000

1000

0

subarray 2

PKA ng/ml

fluor

esce

nce

250 ng/ml Cy3-PKA

Subarray 1

Subarray 2

Cy3-BSA Cy5-BSA Cy3-PKA

2 5 10 20 50 100

500

2000

1000

0

subarray 1

PKA ng/ml

fluor

esce

nce

2 5 10 20 50 100

500

2000

1000

0

subarray 2

PKA ng/ml

fluor

esce

nce

Cy3-PKA

Cy3-PKACy3-BSA

Cy3-BSA

Data analysis

establish: dynamic rangesensitivityspecificityreproducibility

En

erg

y

LUMO

HOMO

Molecule

Bandgap

Nanoparticle

con- ductingband

valenceband

Bulk crystal

Size

h

Nanoparticle fluorescence labeling

Photoluminescence of CdSe nanorods with different width.

400 500 600 700 800

0.0

0.2

0.4

0.6

0.8

1.0

PL

/ a

.u.

/ nm

CdSeCdTe

Tunable fluorescence spectra

Rh

od

amin

eN

ano

rod

sN

ano

rod

sN

ano

rod

sR

ho

dam

ine

Rh

od

amin

e

0 1 2 3 4 5 6 7

05

10

15

20

25

time

fluo

resc

en

ce

+ brigthness

- bleaching

Nanoparticles vs. organic fluorophores

“In comparison with organic dyes such as

rhodamine, this class of luminescent labels is 20

times as bright, 100 times as stable against

photobleaching, and one-third as wide in spectral

linewidth.”

From: W. C. W. Chang, S. Nie, Science, 1998, 281, 2016-2018.

Standardization

Pubmed: gene expression profiling [MeSH] gene expression profiling / (methods OR standards) [MeSH]

number ratio

1998 1999 2000 2001 2002

05

00

10

00

15

00

1998 1999 2000 2001 2002

0.0

00

.05

0.1

00

.15

0.2

0

Five aspects of standardization

control systematic

errors (calibration)

control systematic

errors (calibration)

W. Huber, Div. Molecular Genome Analysis, DKFZ

exchange of models

exchange of models

exchange of data

exchange of data

control stochastic errors

(statistics)

control stochastic errors

(statistics)

publicationpublication

Calibration: control systematic errors

standardized experimental

conditions

(cell lysis, labeling, desalting; array coating spotting buffer, scanner

settings)

standardized experimental

conditions

(cell lysis, labeling, desalting; array coating spotting buffer, scanner

settings)

W. Huber, Div. Molecular Genome Analysis, DKFZ

regression of correction

transformations

(label incorporation rate, absolute amount

of sample, probe sensitivity)

regression of correction

transformations

(label incorporation rate, absolute amount

of sample, probe sensitivity)

Calibration: control systematic errors

W. Huber, Div. Molecular Genome Analysis, DKFZ

ik ik ik iky a b x quantity of

interest (transcript

abundance)

measured quantity

(fluorescence intensity)

Calibration:- establish a useful calibration model- estimate (enough about) aik, bik

- determine error bars

Standardization of data analysis and modeling efforts

Software development in biosciences research:trade-off between o performance, specialisation o platform-independence, common standards, modular architecture fragmentation of languages, data formats, analysis

and modeling platforms inflexible monoliths, becoming harder to maintain

than to abandon- a tendency for empire-building…

"A program that only runs at one site is not software, it’s a piece of hardware" R. Gentleman

Standardization of data analysis and modeling efforts

Approach the writing of academic software like a scientific publication:

o peer-reviewed

o easily available and widely exchanged between scientists

o adherent to standards (formal and informal)

o augmental, modular

Examples:R (Statistics) Bioconductor (Functional Genomics) Bioperl (Human Genome Project)

SBML: systems biology markup language

W. Huber, Div. Molecular Genome Analysis, DKFZ

Standardized exchange of systems biology model descriptions between

different groups / software

MAGE-ML: Microarray and Gene Expression Markup Language

W. Huber, Div. Molecular Genome Analysis, DKFZ

Standardized exchange of experimental data on transcript or protein

abundancesPublic Database

LIMS

Project Database

Institutional database

Publication of systems biology research

In systems biology, results will be published as dynamic documents, mixing data, program code, graphical visualisation, descriptive text.

… WWW "supplements"… Sweave… several research projects are under way (eg “Reproducible Research Project”, Bell Labs / Harvard)

NTI

NTI

NTIII

TIII

NTIII

M C

In traditional biology, results can be published as still image, static diagram, or text:

Thank you