Goal: The course introduces basic concepts of molecular modeling and simulation. We will cover a variety of topics including visualization, molecular mechanics, several different simulation techniques (Molecular Mechanics, Molecular Dynamics etc.), and the analysis of the model structure. At the end, attendees will work through a short tutorial on the topics discussed in the class.

Click here to download the recent class slides (pdf file) and hands-on session notes (pdf file). The data files used in the hands-on exercise can be downloaded from here.

Software: DS Viewer Pro 6.0 (Accelrys Inc.) will be used to do the hands-on exercises.

Click here to go back to my class web-page.

Molecular Modeling Hands-on Exercises using DS ViewerPro 6.0

Molecular Modeling Hands-on Exercises using Sybyl

Use Sybyl (6.9.2) for the following two exercises.

• Exercise 1: Building Aspirin (2-(acetyloxy)benzoic acid) molecule (SMILES script for Aspirin is CC(=O)Oc1ccccc1C(=O)O).

  We will load the pre-built aspirin mol2 (sybyl file type) into sybyl. The pre-built aspirin molecule has been intentionally made to contain some incorrect atom types. In this example we will correct the atom-types, add partial charges, and use molecular mechanics method to energy minimize the molecule. Finally, we will also display the total energy of the minimized aspirin molecule.

  This exercise has been setup to understand the basic assumptions of molecular modeling using a simple system. It will also anser questions such as: What are atom-types? Why do we have to worry about them? How to calculate partial-charges? What are the different methods of calculating them? What is a force-field? Which one to choose? How to carry-out energy minimization? How to calculate the total energy of aspirin and list its different energy components (electrostatic, van der Waals etc.)?

  1. File >> Read >> Choose "File Type:molecule" >> choose "basaspirin.mol2" >> Hit "OK"
     Note: Use DepthCue option to make your molecule look bright. DepthCue option is accessible from the column of menu options located on the left.
  2. View >> Label Atom Name... >> Click "All" >> Hit "OK"
     Note: The lable size can be changed by clicking on the box labelled "Font" located on the left-most of the Sybyl main window.
  3. View >> Label Charges...
  4. View >> Label Atom Type... >> Click "All" >> Hit "OK"
  5. As indicated above some of the atom types of Aspirin are not correct. For example, carbon atoms belonging to aromtic rings should have atom types as C.ar. To fix/correct the atom-types do the following:
  6. Build/Edit >> Modify >> Atom...>> ONLY_TYPE >> hit "OK" >> Click "All" >> Hit "OK". This will select all the atoms and open the "Option" window.
  7. As the green marker moves through the atoms of aspirin, Choose proper atomtypes (see figure for correct atom-types) and hit "OK".
     Aspirin with correct sybyl atom types are shown in the displayed figure
  8. Build >> Edit >> Add >> Hydrogens
  9. Compute >> Charges >> Gasteiger-Huckel >> Anser "No" to the question "Do you want to change formal charges before computing charges?"
  10. Computer >> Minimize... >> Click "Energy Setup: Modify" button >> Charges "Use Current" >> Hit "OK" >> Again hit "OK". Click here to see the minimized structure displayed with a VDW dot surface.
  11. Compute >> Energy >> Hit "Compute" to compute the energy. Type "C" to see the additonal Output.

• Exercise 2: How to align two protein structures using homology?

  For this exercise, we will be aligning two forms of asparaginases (1wsa,1hfw) using the biopolymer module of Sybyl (6.9.2). Click on the respective pdb entries to download the asparaginases.

• 1wsa: Wolinella Succinogenes L-Asparaginase II
• 1hfwAC: (Erwinia Chrysanthemi L-Asparaginase A & C chains only).
1. File >> Read >> 1wsa.pdb >> Hit "OK" >> Anser "No" not to center the molecule.
2. File >> Read >> 1hfwac.mol2. This molecule will be put in M2 area.
3. View >> Color >> Atoms.. >> Choose M1 & Left-Click All >> OK >> Cyan
4. View >> Color >> Atoms.. >> Choose M2 & Left-Click All >> OK >> Yellow Click here to see the figure.
5. Biopolymer >> Display >> C-alpha only >> Choose M1 & hit "OK"
6. Biopolymer >> Display >> C-alpha only >> Choose M2 & hit "OK"
7. Build/Edit >> Delete >> Atom... >> Left-click to select M1 >> Click on "Sets..." >> select "WATER" >> hit "OK"
8. Build/Edit >> Delete >> Atom... >> Left-click to select M2 >> click on "Sets..." >> select "WATER" >> hit "OK"
9. Biopolymer >> Align Structures Using Homology... >> To Specify Fixed protein (highlight m1), and to specify Movable protein (highlight m2). Choose "Calpha" as the type of atoms used for fitting, and click "Align"
10. After a brief break, main window shows the aligned proteins. Also watch the command prompt window to see the distance between each residues with the Weighted Root Mean Square Distance value
• Exercise 3: Protein Visualization using Sybyl. For instructions, click here.

  Use Accelrys InsightII for the following exercise

• Exercise 1: Hydrophobicity calculation, surface creation and visualization using InsightII

Take Home Exercise: Case Study

Aim: In this case study, we will start with a fragment of a DNA sequence which encodes a protein. We will translate the nucleic acid sequence and identify the protein using the standard bioinformatics techniques. Upon identifying the protein, we will extract the whole protein sequence and use the analysis tools to find sequence neighbours. We will also search for patterns, domains and identify any fingerprints in the model sequence. Finally, we will predict the secondary structure and if the structure is unknown, exploit the homology to build the 3-D structure. We will further analyze the protein structure using popular software such as InsightII and Sybyl.

Mystery Model (nucleotide sequence)

aaaaatgaac caataattac aagttcaaaa

Translation

• For the first part of this exercise, we will use the composite database, OWL. Use the analysis tools interface to submit the above nucleic acid query, see Fig 1. The six translated frames are shown here in Fig 2. In the previous figure the stop codons are denoted by the symbol (!). Ignoring the sequences with stop codons, we are left with forward0 and reverse1 amino acid sequences. Which is the correct sequence?

  Identifying the correct fragment

• To start with we will use the forward0 sequence, KNEPIITSSK , and search in the nr (NCBI) database. Since we want to find the exact match, we will choose the Protein Blast and use search for short nearly exact matches option. The results of the search are shown in this figure Fig 3. This concludes that KNEPIITSSK , is the correct protein sequence. To see the results of the blast search for reverse1 (FELVIIGSF), click Fig 4. Similar results were obtained when searched through the OWL composite database.

  Identifying the family & Looking for patterns

• This concludes that the forward0 is the correct partial sequence. From NCBI (NR) search we found that the partial sequence is actually the part of a protein similar to Bacteriophage T4 Lysozyme. Blast is not always guaranteed to give clear-cut results. So, to confirm the results, scan the sequence (pattern) using PROSITE database Fig. 5. To see the results click here (Fig.6). Also from the PFAM database, we identify that this sequence belongs to "Phage Lysozyme" (Accession number: PF00959). Click here for the PFAM reults (Fig. 7). Swiss-Prot also categorises the parent protein for our sequence to be a hypothetical protein meaning that there is no experimental evidence that this protein is expressed in vivo . For more information on the results of the Swiss-Prot search, click here (Fig.8). The full protein sequence in fasta format is shown below

  >tr|Q8T1H7 Hypothetical 19.4 kDa protein - Dictyostelium discoideum (Slime mold).
  MVSSIKDMLKYDEGEKLEMYKDTEGYYTIGIGHLITRIKERNAAILSLEEKIGHKVKMDS
  KNEPIITSSKSEALFEKDLSVATKSIESNPTLSTIYKNLDNIRKMAIINMVFQMGVNNVL
  TFKMSLKLIEEKKWAEAAKEMKNSTWNHQTPNRSNRVISVIETGTLNAYK
  

  Click here (seq.fasta) to download the protein sequence.

  Searching the BLOCKS database

  The URL for BLOCKS database is http://blocks.fhcrc.org/blocks/blocks_search.html . BLOCKS is an automatically generated database of ungapped multiple sequence alignments that correspond to the most conserved regions of proteins. For results, click here.

  Searching the PRINTS database

  PRINTS database defines a fingerprint for protein families. It stresses the fact that a group of motif defines the protein families. The Web-link for PRINTS database is http://www.bioinf.man.ac.uk/dbbrowser/PRINTS/

  To see the search results, click here. For the graphic display of GRAPHScan, click here.

  Homology Modeling: Identifying the template(s)

  Swiss-Model was used to homology model the protein sequence. The results of the Swiss-Model can be obtained by clicking here. The model protein in PDB format can be obtained by clicking here. To visualize the model protein, click here. here The Psi-Blast results show that the closely related 3D-strucure is 1L24, for the final results click here (Fig.11).

  Analysis

  1. Compare the results from Pfam, PRINTS and BLOCKS. Do the results agree with each other?
  2. Is the Swiss-Model OK? Did it predict the 3D structure for all the input amino acid sequences?
  3. Predict the secondary structure using PredictProtein server? Is the secondary structure evaluation agrees with the Swiss-Model built 3D-structure?
  4. Evaluate the quality of the structure. How about using procheck for the structure evaluation?
  5. Calculate some of the important properties like, solvent accessible surface area (SAS), Delphi electrostatic properties, location of hydrophobic and hydrophilic surface patches (if any).

  Modeling Resources

  Visualizer

  1. SPDBV Swiss-PDB Viewer
  2. Cn3D See in 3-D
  3. Rasmol Rasmol
     • Rasmol Rasmol Animations

  Computational Chemistry/Biology Software

  1. Sybyl Tripos (Flexi-Dock, QSAR, Gen-Fold etc.) ($$$$$)
     SYBYL Local information on using SYBYL on ABCC computers.
  2. InsightII Accelrys (GCG, InsightII, Homology, Modeler etc.) ($$$$)
     Insight II Local information on using InsightII on ABCC computers.
  3. AutoDock Automated Docking of Flexible Ligands to Macromolecules (Scripps Research Institute)
  4. DOCK Small Molecule Receptor docking (UCSF)
  5. Amber Assisted Model Building and Energy Refinement. Biomolecular MM and MD simulations.
  6. GROMACS GROMACS: The Fastest Molecular Dynamics Program.
  7. VMD Visual Molecular Dynamics software (NIH)

  Literature Survey

  1. Medline NCBI-Medline

  Databases

  1. GenBank NCBI-Gene Bank
     • GenScan. Gene Scanning and identification.
       • Search GENSCAN using Fragment of Chromosome 1 of Arabidopsis. It contains 102,000 base pairs of DNA.
  2. Proteomics Tools Expassy Proteomics tools (Protein Identification, DNA -> Protein, Similarity Searches, Post-translational modification prediction, Primary/secondary/Tertiary structure prediction and many more)
  3. Protein Machine Translation tools at European Bioinformatics Institute, UK
  4. EMBL European Molecular Biology
  5. European Bioinformatics Institute (EBI)
  6. MIPS EMBnet Special Node, MIPS
  7. BSM Biomolecular and Structure Modelling Group, UCL, UK
  8. Swiss-Prot ExPassy: Access to databases Swiss-Prot, TrEMBL, Prosite, SWISS-2DPAGE, ENZYME, SWISS-MODEL Repository, CD40Lbase, SeqAnalRef from ExPASy home page
  9. TrEMBL TrEMBL Database contains all the coding sequences (CDs) from the EMBL nucleotide sequences.
  10. PIR Protein Information Resource
  11. nrl3d NRL-3D
  12. RESID RESID database is a comprehensive collection of annotations and structures for post-translational modifications.
  13. EMBNET Switzerland Blast, Box-Shade, DotPlot T-COFEE and many more

  Composite Databases

  1. NR All non-redundant GenBank CDS translations+RefSeq Proteins +PDB+SwissProt+PIR+PRF
  2. OWL non-redundant datbase of four publically available sources Swiss-Prot, PIR(1-3), GenBank (translation) and NRL-3D.
  3. UNIPROT Universal Protein Resource [Swiss-Prot, PIR and TrEMBL]
     1. Swiss-Prot+TrEMBL Swiss-Prot Composite Database

  Sequence Comparison

  1. BLAST NCBI Blast
  2. Dotlet Swiss-Institute for Experimental Cancer Research, a founding member of Swiss Institute for Bioinformatics.
  3. Pfam MSA C2H2 zinc finger Domain multiple sequence alignment.

  Secondary Databases

  1. Prosite Database of protein families and domains.
  2. PRINTS Protein FingerPrint Database. Search initiated from BLOCKS database uses PRINTS fingerprint database
  3. BLOCKS Uses INTERPRO database from EMBL
  4. PFAM Protein families database of alignments and Hidden Markov Models
  5. PRODOM Comprehensive protein domain families obtained from Swiss-Prot and TrEMBL.

  Structural Databases

  1. PDB Protein Data Bank
  2. mmCIF Information on macromolecular Crystallographic Information File format
  3. MMDB Molecular Modeling DataBase: A database of 3D structures, as well as tools for their visualization and analysis.
  4. SCOP Structural Classification of Proteins
  5. Prof. Thornton-PDBsum A database of known 3D strucutres of proteins and nucleic acids
  6. PDBREPORT Database reports structural problems in PDB entries using WHAT_CHECK software
  7. CATH Protein Structure Classification
  8. FSSP Fold classification based on Structure-Structure alignment of Proteins

  Protein-Protein/Protein-DNA Interactions

  1. Protein-Protein interaction server
  2. DNA-Protein interaction server

  Secondary Structure Prediction

  1. Predict Protein Server The Predict Protein Server
  2. JPRED Server Barton Group Secondary Structure Prediction Server
  3. PSIPERD Secondary Structure Prediction Server
  4. EVA EVA continously and automatically analyses protein structure prediction servers in 'real time'
  5. SAM-T02 HMM-based Protein Structure Prediction

  Identifying Transmembrane segments

  1. Protscale Expasy web-site for hydrophobicity prediction
  2. TMHMM Predicting transmembrane regions using Hidden Markov Models. This site is maintained by Center for Biological Aanalysis (CBS), the Technical University of Denmark.
  3. PHDhtm PHDhtm TRANSMEMBRANE HELICES PREDICTION

  Homology Modeling

  1. Swiss-Model An Automated Comparative Protein Modeling Server
  2. Gene-Mine Gene Analysis Engine and Viewer for Unix. Formerly called Xlook.
  3. Modeller Program for Comparative Protein Structure Modeling by Satisfaction of Spatial Restraints

  Structure Validation and Quality of protein models

  1. Procheck Stereo-chemical quality of protein and residue by residue analysis in figures
  2. PDBreport Report of PDB files
  3. VADAR Volume, Area, Dihedral Angle Reporter Report of PDB files
  4. DSSP Definition of secondary structure of proteins given a set of 3D coordinates
  5. Verify_3D Verify3D Structure Evaluation Server

  Evolutionary Conservation in 3D protein structures

  1. Consurf Server Server for identification of Functional Regions in Proteins

  Structure Comparison

  1. SuperPose
  2. Deep-View or Swiss PDB Viewer Structure overlay can be carried out using Magic Fit module
  3. Other commercial tools include, Sybyl, InsightII etc.

  Electrostatics and visualization

  1. Delphi Commercial Version of Delphi. A module in InsightII.
  2. Grasp Graphical Representation and Analysis of Structural Properties. A Molecular Visualization and analysis program. It is particularly useful for the display and manipulation of the surfaces of molecules and their electrostatic properties.

  Force-Fields

  1. MM2,MM3,MM4 Force-Field

  Drug Information

  1. Sigma_Aldrich
  2. United States Pharmacopeia

  Biochemical Pathways

  1. Biochemical Pathways aMAZE: Workbench for the representation for the representation, management and analysis of information of cellular processes, genetic biochemical pathways, signal transductions.
  2. Expasy Biochemical Pathways
  3. KEGG Kyoto Encyclopedia of Genes and Genomes

  Other Useful Resources

  1. OMIM Information about disease causing genes.

  Recommended Books/Web-links

  • Molecular Modeling: Basic Principles and Applications, Second Edition, H.-D. Holtje, W. Sippl, D. Rognan, G. Folkers (2003)
  • Sequence Analysis in a Nutshell: A guide to Common Tools and Databases Scott Markel, Kristine Conner, Darryl Leon (2003)
  • Molecular Modeling and Simulation, T. Schlick (2002)
  • Molecular Modeling, A.R. Leach (2001)
  • Boinformatics: A Practical Guide to the Analysis of Genes and Proteins Andreas D. Baxevanis, B. F. Ouellette (2001)
  • Beginning Perl for Bioinformatics James D. Tisdall, Betsy Waliszewski (2001)
  • Bioinformatics Basics: Applications in Biological Sciences and Medicine, H.H. Rashidi and L.K. Buehler (2000)
  • Bioinformatics Computer Skills, Cynthia Gibas and Per Jambeck,(2000)
  • Computer Simulation of Liquids, M.P. Allen & D.J. Tildesley (1989)
  • Fortran 90/95 by S.J. Chapman

  On-line Tutorials

  • A Tutorial on Genes, Genomes, and Bioinformatics Excellent Tutorial.
  • International Union of Pure and Applied Chemistry Recommendations on Organic & Biochemical Nomenclature, Symbols and Terminology etc.

  
  Contact Info: 
  Dr. S. Ravichandran 
  Advanced Biomedical Computing Center, 
  NCI-Frederick, 
  NIH
  Bldg 430, Frederick, MD 21702
  Tel:  301-846-1991 
  Email: sravi at ncifcrf dot gov
  Web:  http://ncisgi.ncifcrf.gov/~ravichas
                  
  
  

  Web-Page Updated on May 8 2008