P3Fold is a protein folding pathway prediction framework that combines ensemble modelling techniques with evolutionary sequence information method
P3Fold inputs consist of i) a single AA sequence, ii) a set of parameters controlling the size and complexity of the conformational landscape to be explored (i.e., the number and length of Β-strands, and the minimum inter-strand loop size). The web-server contemplates three possible scenarios i) Let P3Fold predict topology ii) Provide Topology and iii) Compute all attainable topologies. For the second case (i.e., Provide Topology) the user must input the protein topology as a string. This string should be a permutation of the first M integers and M-1 three possible characters (P, A or N, for Parallel, Antiparallel and None interactions), where M is the number of strands provided by the user. It is important to stress that the integers and characters must be interleaved (e.g., 3A4P1A2) In the case when P3Fold was not able to compute a topology (because there was not enough evolutionary information to make a prediction, or because an internal error), P3Fold would compute the full energy landscape (i.e., third scenario). In the third scenario, P3Fold will compute a network with all the attainable protein topologies. It is important to mention that this scenario is much slower than the scenarios where only one topology is computed.
Once the user enters the input data and click the submit button, P3Fold will run external software to compute evolutionary information (i.e., to compute MSAs and predict secondary structure, residue contacts and Β-strand pairings). User can follow the status of those tasks in the status-html page created by P3Fold. Once the computation is done, the user can click on the link that will depict the results in our protein network visualizer.
The network mode represents entities as round nodes and edges represent the relationships between them. In the ring mode, nodes are localized in the perimeter of a circle in which the radius depends on the number of strands
The user can zoom in the information from a specific topology by typing the desired topology in the search Box or selecting it from the list displayed by the Topology Selection
Show Dynamics: This button allows access to the folding dynamic results. The user will find a subgraph composed by all the conformational states for which a folding trajectory intersects. Nodes are represented by a graphical representation of the topology of the protein ensemble
Pathways: This button will show the predicted folding pathways
Dynamics Movie:The user sees the predicted folding dynamics as a movie of the folding time evolution for the selected protein
>Back to zoom (circles): These buttons allow going back to the initial representation of the energy landscape
The information regarding the current protein ensemble is activated when the user clicks on one of the topologies. Particularly, a box containing information about topological labeling of the ensemble is depicted just beside the clicked node. A graphical representation of the secondary structure conformation through an array of directional arrows is used to indicate the topology of the protein ensemble. The arrows are colored using a blue to purple color scale that represents the position of each strand with respect to the sequence
Shape: It is a string that represents the protein topology
ID: Unique identifier
Probability: Boltzmann probability of the protein ensemble
Residues: Curly brackets represent the indexes of strands. Square brackets gives a starting point of the residue interactions. For example, in a topology 1A2, the line (1,4);(14-16),[14,4] can be interpreted as follows: There are two strands (first strand involves residues 1 to 4, second strand involves residues 14 to 16). The residue interactions are 14-4, 15-3 and 14-2
Button Display Pathways: This button allows the visualization of the reported paths selected by the user
Please remember that if you already have result data from P3Fold you can use our P3Fold Data Visualizer!