Search page
PDB structures
Uploaded structures
Loading page
Results page (Figure 1) (you can find an example output page here)
1) General details about the analysis, with the name of the structure, the link to the results page and a button for the download of the results, containing the PDB structure with the predictions and the results tables.
2) Main menu of the results page. Clicking on one button will show the corresponding table with the results for a particular nucleotide type and will zoom on the top prediction in the jmol visualization applet.
3) Jmol structure visualization applet (takes few seconds to load). Nucleobases are colored in blue, carbohydrate in red, and phosphate in green.
4) Buttons are available below to show/hide any crystallized ligand in the query structure, show/hide the protein surface, zoom in and zoom out (this can be done also with the mouse wheel).
5) Results table for a particular nucleotide type. Predictions (if present) are displayed in rows and sorted by score.
Each row reports the:
The prediction box (see Figure 2) is divided into sections, one for each nucleotide module composing the prediction (displayed with different colors, light blue for the nucleobase, red for the carbohydrate and green for the phosphate). For each module, the section reports:
Downloadable results file
In the first part of the results page you can click on the "Download Results" button; after decompressing the folder you will get:
PDB structures
Insert one or multiple PDB codes in the text area on the left. (upper or lower case).
Multiple codes can be written one per row, or in the same row separated by a comma.
You can specify a chain of the PDB structure to be analyzed by adding the chain letter to the PDB code (ex. 1phkA).
Uploaded structures
You can upload multiple PDB structures at once using the upload button on the right.E-mail
If you want the conservation to be calculated for the residues of the uploaded structures a file named "aliases.txt" must be uploaded. The "aliases.txt" file specifies which PDB structure must be used as reference for an uploaded structure in order to calculate the conservation of its residues. The file must be formatted as follows, with the name of the uploaded structure file, the reference (alias) PDB code to use and the chain of the uploaded structure you want to analyze, all separated by a TAB. An empty alias field means that the conservation will not be calculated, an emtpy chain field means that all the chains in the structure will be analyzed.
Example
Note: you can give both uploaded and PDB codes as input. A quality check of the codes and uploaded structures will show any occurring problem.
1CXZ.pdb 1cxz B
5UKD.pdb 5ukd A
1PHK.pdb 1phk A
You can provide an email address. Nucleos will send an email containing the links to the results page of the submitted structures.Analysis examples
Note: the email is optional, and not required by Nucleos to carry out the analysis.
Three examples of identification of nucleotide-binding sites are provided in the search page.
Loading page
When the submission is complete, a page reporting the real-time status of each analysis will be shown, with the name of the structure, the percentage of the analysis done and a link to the results
Clicking on this link will show the final results page, described below. A page reporting the status of all the steps of the analysis will be shown if the analysis is not completed.
When the analysis is done this page will automatically show the results.
Results page (Figure 1) (you can find an example output page here)
1) General details about the analysis, with the name of the structure, the link to the results page and a button for the download of the results, containing the PDB structure with the predictions and the results tables.
2) Main menu of the results page. Clicking on one button will show the corresponding table with the results for a particular nucleotide type and will zoom on the top prediction in the jmol visualization applet.
3) Jmol structure visualization applet (takes few seconds to load). Nucleobases are colored in blue, carbohydrate in red, and phosphate in green.
4) Buttons are available below to show/hide any crystallized ligand in the query structure, show/hide the protein surface, zoom in and zoom out (this can be done also with the mouse wheel).
5) Results table for a particular nucleotide type. Predictions (if present) are displayed in rows and sorted by score.
Each row reports the:
- rank of the prediction6) Jmol checkboxes to show/hide all nucleobases, carbohydrate and/or phosphates predicted on the query structure.
- score (reported also as the sum of the two terms, the cluster score and the conservation score, for more information on how to interpret the score go to the Usage page)
- architecture of the prediction (the "N" "C" and "P" letters represent the nucleobase, the carbohydrate and the phosphate respectively)
- display checkbox to highlight/hide the prediction in the jmol applet.
- expand/close button to show/hide additional information on the prediction (described below).
Figure 1
The prediction box (see Figure 2) is divided into sections, one for each nucleotide module composing the prediction (displayed with different colors, light blue for the nucleobase, red for the carbohydrate and green for the phosphate). For each module, the section reports:
1) the structural match of the prediction (reported with the sequence identity between the chain of the query protein and the chain containing the template binding site), with the amino acids of the query structure matching amino acids of a template protein (structure name, protein chain, name and position of the matching residues are reported).
2) a "Highlight residues" checkbox to highlight/hide the amino acids of the query structure in the jmol applet, coloring them with the appropriate color.
3) a button that allows the visualization of the template binding sites for that nucleotide module in its original structure; clicking on the button opens a new small-sized browser page with a jmol applet zooming on the nucleotide module and corresponding template binding site.
Figure 2
Downloadable results file
In the first part of the results page you can click on the "Download Results" button; after decompressing the folder you will get:
1) Four PDB files with the extension ".NCP.combinations", ".NCPP.combinations", ".NCPPP.combinations" and "NCPPCN.combinations" containing the query structure with the predictions for that particular nucleotide type, respectively the AMP-like, ADP-like, ATP-like and NAD-like predictions.
2) A PDB file with the extension ".predictions_ncp" containing all the predicted nucleotide modules, before nucleotide-binding sites reconstruction.
3) A results table (text file with tab-separated columns) with the extension ".single_prediction_details" containing the information about the predicted nucleotide modules before their combination reporting their
type (e.g. nucleobase)
name in the ".predictions_ncp" PDB file
distance from the surface (in Å)
cluster score
conservation score
chain and residues of the query protein involved in the prediction
chain and residues of the protein with the template binding sites involved in the prediction.
4) A results table (text file with tab-separated columns) with the extension ".combined_predictions_details" containing the information about the reconstructed nucleotide-binding sites of different types. This file reports the
prediction type (e.g. NCP stands for nucleobase-carbohydrate-phosphate, an AMP-like prediction)
rank
score
Nucleotide modules used to build the complete nucleotide-binding site (or part of it if a complete reconstruction is not possible).