A New Strategy of Protein Identification in Proteomics

1
A New Strategy of Protein Identification in
Proteomics
Xinmin Yin CS Dept. Ball State Univ.
2
Contents

1. Background.
2. Problems.
3. Proposal.
4. Conclusion.

3
Whats Proteomics ?
Use of quantitative protein-level measurement of
gene expression to characterize biological
process and to decipher the mechanisms of gene
expression control.
4
Whats Proteomics ?
DNA
transcription
RNA
translation
PROTEIN
5
Current Research in Proteomics
Quantitative measurement of proteins expressed in
a cell protein identification protein
quantification
6
Experimental Flow Chart of Protein Identification
cell
mass sequence
proteins peptides
data processing
7
Schematic of MALDI MS/MS
Laser

Detector1 E1 MS1 E2
Peptides
MS2
Detector2
Matrix Assisted Laser-induced Desorption
Ionization
8
An Example of MALDI MS/MS Spectrum
9
Problem of Scaling up MALDI MS/MS to Proteomics
Number of peptides is huge Human genome
(60,000 proteins) 2.6 X 106
peptides
It will take THREE MONTHS to identify all
proteins in a human cell.
10
My Proposal
To develop a new data analysis method

• Avoid using MS2 if a peptide can be identified
  in MS1.
• Reduce size of peptide database during data
  analysis

11
Facts of MALDI MS/MS Experiment

• 40 peptides are unique in mass when mass
  accuracy of MALDI is better than 10ppm.
• Analyzing MS1 spectrum is much easier than
  analyzing MS2 spectrum.
• It takes about 1 to 2 min for analysis of a
  single MS2 spectrum through database.

12
My Proposal

• Using local protein database for data analysis.
• Developing a peptide filter.
• Automating data transfer from MS1 to MS2

13
Local Protein Database
PURPOSE 1. Make peptide database small
according to experimental methods 2. Do
online data processing

• Download protein database from public sites.
• Digest proteins in computer.
• Select peptides according experimental methods,
  and store them in memory.
• Calculate peptide chemical modification.

14
Generating Peptide Database
Protein Database
Peptide Database
Enzyme Digestion
Each protein can produce 30 100 peptides
15
Peptide Filter
PURPOSE 1. Do online data processing 2.
Reduce peptides database size 2. Reduce number
of peptides passed to MS2

• Identify peptide by mass alone.
• Sort identified peptides according to proteins.
• Delete rest peptides of the identified protein
  from peptide database.
• Pass unidentified mass information to MS2.

16
Identifying Peptide by Mass Alone
Mth
Mexp
Npep
?M


?M Mth Mexp If (?M lt ?) Npep
Npep number of mass matched peptides ? mass
accuracy of MS1
17
Reducing Size of Peptide Database
If (Npep 1) Peptide is unique
Deleting peptides from the same protein
Else Send peptide to MS2 for sequence
information.
18
F L O W C H A R T
19
Program Design
Computer Language C
Classes
Protein
Peptide
Linked list
20
Conclusion
1. My program will generate a list of peptides
from the protein database. 2. Peptide
identification is carried out while doing
experiment. 3. Data analysis time will be
significantly reduced. 4. This program will
make identification of proteins from a cell
possible and reliable.