        |
 |
|
|||
|
|
 |
| Home | Folding Submission Form | Pathway Submission Form | Folding Results Database | FAQ |
A roadmap is what we call the graph approximating the protein's potential landscape that is generated by our technique. Please see the Protein Folding Server Technique FAQ or our research project webpages for more information about our technique and research.
The roadmap name forms the first component of the filename for all outputs related to the roadmap:
| Format: Roadmap names are of the form "Ri-NCID", where i is a unique numeric identifier associated with each roadmap generated by the server, and NCID is the PDB ID of the protein's native conformation. |
| Example: Roadmap 14 for protein A, which has PDB ID "1BDD", would be referred to as "R14-1BDD". |
This plots the backbone's phi vs. psi torsional angles (of every residue) for all conformations in the roadmap. As seen in the figure below, clustering occurs in areas of stability such as alpha helices and beta sheets.
| Format: File names are of the form "RoadmapName.phi-psi.ext", where ext is either gif, pdf, or ps. |
| Example: The plot for roadmap R14-1BDD in pdf format would be named "R14-1BDD.phi-psi.pdf". |
| Classic Ramachandran Plot (from www.cryst.bbk.ac.uk) |
Roadmap Phi/Psi Plot (Protein A, all alpha) |
Roadmap Phi/Psi Plot (Protein G, alpha & beta) |
|---|---|---|
 |
 |
 |
This plots the potential energy vs. the root mean square distance (RMSD) to the first target conformation (usually the native state) for each conformation in the roadmap.
| Format: File names are of the form "RoadmapName.pot-rmsd.ext", where ext is either gif, pdf, or ps. |
| Example: The potential vs. RMSD plot for roadmap R14-1BDD in pdf format would be named "R14-1BDD.pot-rmsd.pdf". |
We have observed that the plots we generate from roadmaps of different types of proteins (e.g., all alpha or all beta) have different shapes, and these might provide some insight into folding behavior. For example, all alpha proteins tend to form individual helices first and then pack together to form the tertiary shape. This is reflected in our plots in the narrow tail region where energy changes very little as the RMSD approaches zero. On the other hand, all beta proteins tend to form secondary and tertiary structure at the same time, which is reflected by smoother plots.
| Protein A (all alpha) | Protein G (alpha & beta) | SH3 Domain from Human Fyn (all beta) |
|---|---|---|
 |
 |
 |
|
The roadmap contains many (typically thousands) of pathways to the first target conformation (usually the native state). The SSFO classification for a roadmap analyzes each of the shortest pathways in the roadmap from unstructured conformations to the first target conformation and determines the (temporal) order in which the secondary structural components (e.g., helices, beta sheets, etc.). Pathways are then grouped according to these formation orders. The SSFO classification shows the percentage of the roadmap pathways that have each distinct formation order. Note that this analysis provides only temporal information, but does not provide actual times.
There are various methods to determine the temporal ordering of secondary structure elements along a given pathway. One method examines the contacts present for each piece of secondary structure at each pathstep. The secondary structure element is said to be "formed" when at least x% of the native contacts are present. We provide results for x = 20%, 40%, 60%, 80% and 100%. We have found that 80% is typically a good value and most of the individual pathway analysis results in the Protein Folding Server database use this value. If you would like your protein analyzed using a different value, please contact us at proteinfolding@cs.tamu.edu.
Another method uses rigidity analysis to determine if secondary structure elements are formed or not at each pathstep. The secondary structure element is said to be "formed" when the (normalized) rigidity similarity from the element in the current conformation to the element in the native conformation is at least x% percent. Again, we provide results for x = 20%, 40%, 60%, 80% and 100%. We have found that 80% is typically a good value and most of the individual pathway analysis results in the Protein Folding Server database use this value.
| Format: File names are of the form "RoadmapName.XSSFO.ext", where X is either empty, "ac", or "hc" indicating the analysis considered rigidity similarity (default (default)), all native contacts, or only hydrophobic native contacts, respectively, and ext is either gif, pdf, ps, tex (a latex table), or txt. |
| Example: The SSFO classification table for roadmap R14-1BDD which considered all native contacts in pdf format would be named "R14-1BDD.acSSFO.pdf". |
| Example: The SSFO classification table for roadmap R14-1BDD which considered rigidity similarity in gif format would be named "R14-1BDD.SSFO.gif". |
Consider the example shown below for protein G (2GB2). When a secondary structure element was considered formed when 80% of the contacts were present, there were three formation orders observed in the paths in the roadmap (the 80% column):
This is consistent with the experimental evidence that the beta 3-4 hairpin forms before the beta 1-2 hairpin.
|
For any given conformation, rigidity analysis can label portions of the protein as rigid, independently flexible, or dependently flexible. A rigidity map shows how pairs of residues in a given conformation relate to each other in terms of their rigidity labeling. In a rigidity map plot, residue pairs that are both rigid are colored black and residue pairs that are both dependently flexible are colored green. Black and green areas tend to cluster together on the map and typically align with secondary structure elements. A cluster map is similar to a rigidity map, except that it only colors rigid residue pairs black. All other residue pairs are left uncolored.
| Format: File names are of the form "RoadmapName.XMap.ext", where X is either "Rigidity" or "Cluster" indicating the type of coloring used, and ext is either gif, pdf, or ps. |
| Example: The rigidity map plot of the first target conformation for roadmap R14-1BDD in pdf format would be named "R14-1BDD.RigidityMap.pdf". |
| Example: The cluster map plot of the first target conformation for roadmap R14-1BDD in pdf format would be named "R14-1BDD.ClusterMap.pdf". |
Rigidity maps and cluster maps can give insight into folding behavior. For example, more rigid portions may be more stable and stay folded longer while more flexible portions may denature quickly. Consider proteins G and L which have similar tertiary structure but different folding behaviors where the order of hairpin formation between the two is flipped. This flipping is reflected in their rigidity maps and cluster maps with a flipping of the rigidity relationships of the hairpins.
| Protein G | Protein L |
|---|---|
| Rigidity Map of the Native Conformation | Rigidity Map of the Native Conformation |
 |
 |
| Cluster Map of the Native Conformation | Cluster Map of the Native Conformation |
 |
 |
|
Population kinetics provides information about the time evolution of different conformational populations. We use the Map-based Monte Carlo (MMC) method to stocastically extract pathways and compute population kinetics. The population kinetics plots on our folding server show the time evolution of two sets of conformations: folded and unfolded. As time progresses (larger time step), we expect the population of the folded state to rise and the unfolded state to drop. An example is shown below.
|
MMC Population Kinetics: Protein G
(Protein structure is shown inset) |
 |
|
MMC provides us a time-based view of the folding process. Within this ordered view, we can examine the formation of certain properties. One of these properties, Helix formation, can be measured as a function of the time step from the MMC simulation. As time progresses in the MMC simulation (larger time step), we expect proteins to become folded, thus this value will maximize. An example is shown below.
| Helix Structure Formation: Protein G | |||||||||||||||||||||||||||||||||||||||||||||
 |
|
MMC provides us a time-based view of the folding process. Within this ordered view, we can examine the formation of certain properties. One of these properties, formation of structure around tryptophan residues, can be measured as a function of the time step from the MMC simulation. As time progresses in the MMC simulation (larger time step), we expect proteins to become folded, thus this value will maximize. An example is shown below.
| Tryptophan Structure Formation: Protein G | ||||||||||||||||||||||||||||||||||||||||
 |
|
We can compute relative hydrogen exchange rates of a set of pathways in the roadmap. This set of pathways may be determined from shortest paths, Map-Based Monte Carlo simulations, or other techniques. (See path extraction) for more details.) We can then approximate each residues hydrogen exchange rate by its relative rigidity or flexibility along the pathway. A relative exchange rates plot shows this normalized approximation of hydrogen exchange for each residue. Typically, exchange rates are lower in areas containing secondary structure and higher in random coils.
| Format: File names are of the form "RoadmapName.exchange.ext", where ext is either gif, pdf, or ps. |
| Example: The relative exchange rates plot of a set of extracted pathways for roadmap R14-1BDD in pdf format would be named "R14-1BDD.exchange.pdf". |
| Relative Exchange Rates Plot (Protein G) |
|---|
 |
|
Individual Pathway Analysis Results
A pathway is a sequence of conformations extracted from the roadmap. The sequence corresponds to a set of roadmap edges that connects the start & final conformations. For a discussion on how pathways are extracted, please visit our technique FAQ.
The pathway name forms the second component of the file name for all outputs related to the pathway: RoadmapName.PathwayName.ResultType.ext. Results are typically provided for two types of pathways: pathways between pairs of target conformations and representative pathways for each distinct secondary structure formation order found in the roadmap.
Format: The PathwayName depends on the type of pathway:
|
| Example: The most dominant SSFO pathway for roadmap 350 for protein 1CFD when considering secondary structure formation occuring at 80% hydrophobic native contacts present would be named "R350-1CFD.hcSSFO80-0". The second most dominant SSFO pathway would be named "R350-1CFD.hcSSFO80-1". |
| Example: The pathway from conformation "1CLL" to conformation "1CFD" in roadmap 350 for protein 1CFD would be named "R350-1CFD.1CLL-1CFD". |
The start and final conformations are the first and last conformations, respectively, of a specific pathway. We provide gif images and pdb files for both conformations. Images of PDB conformations are generated by MolScript.
| Format: The Protein Folding Server generates gif images and pdb files of the start and final conformations. The files are named RoadmapName.PathwayName.X.ext or RoadmapName.PathwayName.X-cID.ext where X is either "start" or "final", cID is the PDB of the conformation (if it has one), and ext is either "gif" or "pdb". |
| Example: The gif version of the start and final conformations for the pathway "R350-1CFD.hcSSFO80-0" would be "R350-1CFD.hcSSFO80-0.start.gif" and "R350-1CFD.hcSSFO80-0.final-1CFD.gif", respectively. |
A pathway profile plots some property of the pathway conformations along the y-axis for each pathstep (displayed along the x-axis). Profiles for SSFO-based pathways start from unfolded conformations and finish at the folded conformation. Profiles include intermediate conformations along the roadmap edges to decrease the distance between any two consecutive conformations to a fixed resolution.
Properties include the potential energy, the internal degrees of freedom present as computed from rigidity analysis, distance measures between the pathstep and the first target conformation (typically the native state), relative distance measures between the current pathstep and the previous pathstep, and contacts present (either all or hydrophobic only). We use the following distance metrics for pathway profiles: RMSD, rigidity distance, and cluster distance. Rigidity distance and cluster distance are two ways to measure the similarity in rigidity/flexibility between conformations as computed by rigidity analysis.
| Format: Files are named RoadmapName.PathwayName.XProfile.ext where X describes the profile type and ext is either "gif", "pdf", "ps", or "m" (matlab data). Profile types include "Energy", "Dof", "RMSD", "RelativeRMSD", "RigidityDistance", "Relative RigidityDistance", "ClusterDistance", "Relative ClusterDistance", and "Contact". |
| Example: The pdf version of the energy profile for pathway "R350-1CFD.hcSSFO80-0" would be "R350-1CFD.hcSSFO80-0.EnergyProfile.pdf". |
Here are some examples of different profiles of the most dominant SSFO pathway determined by rigidity analysis for protein G:
| Energy Profile | DOF Profile | RMSD Profile | Hydrophobic Contact Profile |
|---|---|---|---|
|
|
|
|
| ||
|
A timed contact map shows when each native contact last forms on a pathway. A native contact is a contact between two amino acid residues, that are not adjacent in the sequence, but whose Calpha atoms are less than 7A apart in the protein's native state. A contact map is a triangular matrix which identifies all the protein's native contacts. Both axises represent residue numbers, and there is a mark in entry (i,j) if there is a native contact between residues i and j. The contact map can be constructed considering all native contacts ("ac"), or only hydrophobic native contacts ("hc").
| Format: Files are named RoadmapName.PathwayName.ContactMap-X.ext where X is either "ac" or "hc" indicating if all native or only hydropobic native contacts are considered and ext is either "gif", "pdf", "ps", or "txt" (plain text). |
| Example: The gif version of the all contact map for pathway "R350-1CFD.hcSSFO80-0" would be "R350-1CFD.hcSSFO80-0.ContactMap-ac.gif". |
For example, below is the timed contact map for Protein A (1BDD) for one pathway. Note, this only gives a temporal ordering; the time steps do not reflect actual times. Secondary structure elements are indicated in boxes for reference.
|
We provide animations (movies) of each pathway in mpeg and gif formats. We also provide the PDB files for all conformations in the animations. The animations include intermediate conformations along the roadmap edges to decrease the distance between any two consecutive conformations to a fixed resolution. The total number of frames in the animation is shown on the webpage. Images of PDB conformations are generated by MolScript.
| Format: Files are named RoadmapName.PathwayName.ext where ext is eigher "mpeg", "gif" (animated gif), or "pdbs.tar.gz" (gzipped tarfile containing PDB files). |
| Example: The mpeg animation of pathway "R350-1CFD.hcSSFO80-0" would be "R350-1CFD.hcSSFO80-0.mpeg". |
| If you have any questions or comments, please email proteinfolding@cs.tamu.edu. |
Parasol Home | Research | People | General info | Seminars | Resources Parasol Lab, 301 Harvey R. Bright Bldg, 3112 TAMU, College Station, TX 77843-3112 Contact Webmaster Phone 979.458.0722 Fax 979.458.0718 
Department of Computer Science and Engineering | Dwight Look College of Engineering | Texas A&M University Privacy statement: Computer Science and Engineering Engineering TAMU |