Protein Modeling Workshop at UC Riverside (Nov. 8-9, 2002)
The goal of this workshop is to provide practical
knowledge about protein modeling algorithms, tools and databases. Participants
will be given hands‑on experience with sequence structure and domain
databases, protein sequence and structure similarity searches, and protein
modeling and docking. All software used in the Workshop is available as public
Web servers or can be freely downloaded for academic use.
The lecture will follow a logical path from the
basic notions used in description of protein sequence and structure, through
the concept of sequence similarity, sequence‑structure relationship and
similarities between protein structures, to protein‑protein interactions
and drug design. Computer training will follow the parallel path from usage of
the databases of protein sequences and structures, secondary structure
prediction through sequence similarity searches to homology modeling and
protein-protein docking. The workshop will be based in part on the “Principles of
Protein Structure, Comparative Protein Modeling and Visualization” (http://www.expasy.org/swissmod/course/course-index.htm)
-
Lecture:
o
Description of protein sequence and structure: protein sequence, secondary structure, backbone conformation,
super-secondary structure, tertiary protein structure, folds, domains,
quaternary structure.
o
Similarity between protein sequences ‑ concepts and algorithms.
o
Similarity between protein structures ‑ concepts and algorithms.
o
Relationship between protein sequence and structure – protein structure
prediction.
-
Computer training:
o
Sequence databases (NCBI, Genebank, SwissProt)
§
Looking for sequences of interest
o
Sequence similarity searches with Blast and Psi-Blast
§
Predicting function of hypothetical proteins (Psi-blast)
§
Looking for new genes in the genomic data, verifying results
o
Sequence alignment and HMM databases (Pfam, Superfamily)
§
Predicting function of protein impossible to annotate with Psi‑blast
o
Threading methods (FFAS, Metaserver)
§
Predicting fold and function of the protein impossible to predict with Psi‑blast
and HMM algorithms
o
Comparing alignments obtained with different methods (Blast, Psi‑blast,
HMMs, FFAS)
o
Viewing protein 3D structures with Chime, SwissPdbView and Cn3D
§
Structure
analysis (phosphorylation of tyrosine residue in dual-specificity phosphatases)
(MMDB, CDD, Cn3D)
o
Protein structural databases (Pdb, Scop, Dali, Cath)
§
Looking for proteins of interest in structural databases
§
Comparing recently solved protein structure with known protein structures
(Dali)
o
Secondary structure prediction (Jpred)
§
Verifying distant homology predictions
o
Structure similarity searches (CE, Dali)
§
Comparison of two structures with no detectable sequence similarity with
(CE)
-
Lecture:
o
“Significant sequence similarity usually implicates structure similarity” – the basis of homology
modeling
o
Alignment accuracy and accuracy of models based on homology
o
Homology modeling algorithms
o
Side-chain rotamers and loop building
o
Possible applications of protein models of different accuracy
o
Problems of docking and rational drug design
-
Computer training:
o
Basic protein structure analysis with SwissPdbViewer:
§
Analysis of the active site
§
Building Loops
§
Building a functional unit from a monomer
o
Hands on work on specific modeling topics using the SwissPDBViewer (by
Dimitrios Morikis)
o
Presentation of research examples on protein‑protein interactions (by Dimitrios Morikis)
o
Verification of modeling methods with test cases (proteins with known
structures), checking structural similarity with (CE and
SwissPdbViewer).
o
Protein‑protein docking (GRAMM).
WWW adresses of databases
and tools used in the course:
Software, that can be
downloaded and installed before the course: