The bNAber database is a part of the newly established multicenter CHAVI-ID project, based at The Scripps Research Institute.
bNAber collects detailed information on all known HIV-1 bNAbs, including raw neutralization data from TZM-bl assay, sequences of bNAbs and activity against alanine mutants, predicted epitopes and, when available, 3D structures of bNAbs and their complexes.
Data is available to the user using series of user friendly interfaces that were designed to support typical search strategies. Users have multiple ways of selecting antibodies for comparative analysis, including recognition site (gp160 epitope), donor, breadth or depth of neutralization (high or low). Presenting data for a selected group of bNAbs provides a convenient way to understand the antibody properties important to immunogen design.
The 3D information, including precalculated structure comparisons of bNAbs, is presented using a novel, multi-window display of protein complexes, providing the researchers ways to explore the binding modes of the various antibodies and compare different conformations of the same antibody.
Currently the information in the database is manually extracted from literature, including supplemental materials if available and additional information is automatically downloaded from the LANL HIV Molecular Immunology Database, PDB and GenBank.
The database home page provides an interface to access all of the database content and multiple ways for antibody selection. On the home page, a user can select antibodies for further analysis by different criteria including donor, study, neutralization breadth (% of strains neutralized with IC50 or IC80 < 50 μg/ml) and neutralization depth (median IC50 (μg/ml) against viruses neutralized with an IC50 < 50 μg/ml). Selection by epitope can be done by clicking on the HIV spike protein model (see Figure 1 below).
|Figure 1. The primary access mode to the bNAber database, from selection of a subset of antibodies on the home page to the summary page for the selected antibodies to the dedicated antibody page|
After selection, clicking the “submit” button opens up a summary page of selected antibodies and a pull down menu allowing the user to select further actions. On the summary page, clicking on underlined names of specific antibodies opens individual pages with full information about each antibody in the database (see Figure 1 for a sample series of steps through the main pages of the database). Ab details page combines data on neutralization efficiency (median IC50 (μg/ml) against viruses neutralized with an IC50 < 50 μg/ml) from several studies. In addition, links with PDB codes of coordinates (when available) provide access to a special interface for analysis of antibody-antigen complexes, with tools to identify interfaces between them.
Actions highlighted in the pull down menu from the summary page are available in the current bNAber release, others are in development and the date of their availability is indicated on the menu.
Example of the action “show sequences and download FASTA file”:
Selecting multiple antibodies ‘12A21’, ‘2F5’ and ‘2G12’ from “Select Ab” drop menu, “--any--“ from “Select donor”, “Select IC50 breadth”, “Select IC80 breadth”, and “Select neutralization depth” menus (as shown below),
and hitting “Submit” gives the following output:
From the “Search Results” obtained, the user can either select ‘all’ the antibodies (by clicking the “Select” button at the top left), or select individual antibodies for any of the following queries from the “-- make selection --“ drop menu:
1. “Show sequences and download fasta file”, 2. “Overlay structure (based on Ab, on gp120 or CD4 (rare)) – POSA (select 2 or more antibodies to compare)” or 3. “Display IC50 neutralization profile”
Selecting antibodies – ‘2F5’ and ‘2G12’ followed by selecting “Show sequences and download fasta file”, from the “-- make selection --“ drop menu, gives the following output:
The user can then download the following sequences: 1. fasta light chain AA (amino acid) sequence, 2. fasta light chain DNA sequence, 3. fasta heavy chain AA sequence, and 4. fasta heavy chain DNA sequence.
The action “overlay structure” links to a POSA server for multiple structure analysis, which is discussed below. Having selected and downloaded sequences users can perform sequence alignment by tools such as Jalview (http://www.jalview.org) and correlate Abs sequence similarity with their neutralizing properties (e.g., clustered heatmaps obtained from the next tab, as described below).
This tab provides access to an interface for clustering of the neutralization data that is based on the similarity of neutralization IC50 values against individual virus strains. This tool can help a user answer questions such as comparing neutralization data for several Abs to determine differences and similarities in their behavior relative to particular virus clade or strain, compare neutralization data obtained for different studies or compare and correlate grouping of bNAbs based on neutralization properties and on sequence similarity, epitope or other features.
Neutralization data can be presented in two forms: numerical colored table and heatmaps. Colored tables can be often found in papers describing neutralization data of several antibodies. We implemented this function in bNAber. Browsing large neutralization tables require scrolling through the page. We created a heatmap view of neutralization table that allows large datasets to be easily viewed on one screen (Figure 2). By providing hierarchical clustering we make available another useful feature that helps researchers to show groups, similarities and differences in the neutralization data of several antibodies tested on ~100-200 HIV strains (Figure 2). The log10 values of the IC50 neutralization data for a subset of virus strains (utilizing an IC50 of 50 µg/ml for values designated as >50 µg/ml) is provided to the function as a matrix, and the heatmap is generated using the Euclidean distance and average-linkage clustering method.
|Figure 2. Heatmap neutralization data for selected bNAbs (IC50 in µg/ml) shown in alphabetical order (top panel) and ordered by hierarchical biclustering of the IC50 neutralization data (bottom panel) for the same subset of virus strains (log10 of IC50 data were used, with IC50 >50 µg/ml replaced by 50). See on-line supplement for references to tools used to create this images.|
This view was implemented using the R Enhanced Heat Map (heatmap.2; http://hosho.ees.hokudai.ac.jp/~kubo/Rdoc/library/gplots/html/heatmap.2....) function, incorporating the RColorBrewer (http://cran.r-project.org/web/packages/RColorBrewer/index.html) and gplot (http://cran.r-project.org/web/packages/gplots) libraries.
An integrated Jmol structure viewer allows researchers to view complex structures, including precalculated antibody-antigen binding sites, in detail by providing separate windows for entire PDB structures and individual chains.
Papers describing 3D structures of bNAbs present many details of structural organization of Ab-ENV complexes. The lack of interactivity of these images requires readers go to the PDB database or download the structure coordinates and investigate the structure with visualization tools. To help researchers do this analysis faster and easier we implemented a new multi window structure viewer, developed in Jmol. The viewer presents the complex structure in one window and all the subunits of the complex in smaller windows. Changing between synchronized and independent rotation of the subunits in all the windows allows user to select the direction of the view and magnification that fits his need. Checkboxes allow researchers to easily turn off and on any non-protein molecules (HETATMs such as carbohydrates and ions) that are not water. Contact residues between a bNAb and its antigen are recomputed and emphasized the structure graphs. Figures 3 through 5 show the structure of complex 3SE9 (CHAIN G: ENVELOPE GLYCOPROTEIN GP120; H: HEAVY CHAIN OF ANTIBODY VRC-PG04; L: LIGHT CHAIN OF ANTIBODY VRC-PG04). It can be seen that Ig heavy chain has much larger contact area with Ag compared to the light chain. The lists of residues participating in the contacts can be downloaded for further analysis and use in other programs.
Figure 3 (above): The top portion of the structure viewer contains a large view of the full PDB structure including header information on the color-coding of the different chains present. In this image the Show Carbohydrates and Ligands checkbox is checked allowing us to see non-protein elements other than water (depicted in white).
Figure 4 (above): Structure views containing antigens depict individual protein chains in separate Jmol windows in the central portion of a structure view window. Contact points are depicted as spheres in the color of the contacted chain. In the example below we can clearly see that the bNAbs heavy chain (blue) has more contact points with the GP120 antigen (red) than the light chain (green) does.
Figure 5 (above): Structures with pre-calculated contact points between bNAbs and antigen will contain a Contact Points table at the bottom of the structure viewer. In this example (3SE9) only a small portion of an extensive table is depicted showing individual contact points between chain G (the GP120 antigen) and chain H (the bNAb’s heavy chain).
Structure files of co crystals (antibody heavy and light chains, antigen) were obtained from the PDB and are depicted on bNAber using Jmol. For each co-crystal structure a large window is provided showing all chains in separate colors. The individual chains are also depicted in smaller individual Jmol windows, facilitating the closer examination of potential binding sites. Contact residues between chains of the co-crystal structure were defined as residues whose C-α atoms are in a proximity of less than 7.5 Å between bNAb and Ag molecules. Contact residues are displayed as spheres in the Jmol structure viewers and listed in the Contact Points table on the bottom of any structure page containing pre-calculated contacts. Structures depicting bNAb molecules without any antigen present contain no pre-computed contacts table or multi-window view.
This tab brings the user to the POSA server page with precalculated results of structure alignments of all pairs of HIV bNAb structures and an interface to perform further analysis. POSA server for multiple protein structure alignment by partial order graphs was developed previously in our group and provides several tools and interfaces to study structural similarity and diversity between homologous proteins. The superposition is driven by the structural alignment of selected chains – it can be an antigen or Heavy or Light chain of an antibody. At the next step the rest of the pdb chains are superposed and root mean square deviation of C-alpha atoms is calculated for all equivalent atoms. This approach can identify even subtle differences in the binding mode of the antibody and the antigen. The interface presents the aligned sequences and structures in multiple windows, making it easy for users to analyze and compare the structures of antibody-antigen complexes. Independently, access to the precalculated results and to the POSA interface is available from a pull down menu on the summary pages available from the home page followed by manual selection of specific coordinate sets (see the discussion above). Pairs and groups (up to 5) of antibodies can be selected and their multiple structure alignment can be analyzed by the POSA visual interface. Structure comparisons are presented using a novel, multi-window display of protein complexes. Figure 6 shows an example of POSA structure comparisons of bNAbs complexes PDB:2NY7 and PDB:3u7y aligned on the antigen structure.
Figure 6. POSA structural superposition of two antibody-antigen complexes, 3u7y (NIH45-46 Fab in complex with gp120 of 93TH057 HIV-1) and 2NY7 (Broadly Neutralizing CD4-Binding-Site Antibody b12 and gp120) with primary alignment on the antigen gp120 chains. The superposition illustrates the difference between the two antibodies bound to the same epitope, but at different angles relative to the gp120 surface. Separate windows below allow for separate viewing of individual antibodies.POSA Structure Display Examples:
There are 2 frames on POSA structure display page (see example). The top frame shows the corresponding protein sequence information for the protein structures superimposed. The bottom frame shows the three-dimensional superposition of proteins. The protein sequence information in the top frame could be protein sequence (example) or protein structure alignment in text format (example). In the bottom frame, there is one large structure window (with two or more overlaid structures) and several smaller structure windows (with single protein display). The larger window includes structures shown in all smaller windows. Presenting of structures on the large window could be turned on and off. Structures in all window viewers can be synchronized by checking the "Synchronize" box in the bottom of the frame.
The PDB file names and chain Ids in the structure windows and in the top frame are named correspondingly. The PDB files are named as "user's initial PDB file name or PDB id", concatenated with "chain Id used in structure superposition" and "segments id" (if user selected some segments for superposition, this is optional).
Structure comparison viewer - POSA
Example 1: Two broadly neutralizing antibodies bound to the same region of gp120 but with different angles:
|PDB ID||Broadly neutralizing antibody||Antigen|
|2NY7||b12, Chains H, L - heavy and light chains (H and L)||ENVELOPE GLYCOPROTEIN GP120||3U7Y||NIH45-46, Chains H, L - HEAVY CHAIN, IG GAMMA-1 CHAIN C REGION; LIGHT CHAIN, IG KAPPA CHAIN C REGION||ENVELOPE GLYCOPROTEIN GP160; FRAGMENT:GP120 CORE (UNP RESIDUES 43-122, 201-303, 325-486);|
In this example, the Heavy chains from 2NY7 and 4JDT are superimposed while Light chains and gp120 chains are added later for structure display. As we can see the Heavy chains and the Light chains are superimposed well but the antigen chain are not.
Now we used the antigen chains for structure superposition and added the antibodies (H, L chains) afterwards (at here). You can see that now the antibodies are not aligned – they bound to the same region of the antigen with slightly different angles.
This tab provides an interface for easy input of data (Figure 4) on a new bNAb. This online form allows the user to submit the defining characteristics of a new bNAb, such as name, donor, journal, germline sequence, etc. For large-scale studies we invite researchers to submit data tables (as a file attachment through the provided interface), which we will upload into the database. Before submission, the form will be validated for required fields as well as for expected values (e.g. neutralization data should be non-negative). Once the form has been validated, the submitted data is put into a temporary holding area, which is periodically validated by the annotator. Once the annotator has validated that the data is not spam, the data is sanitized against MySQL injection attacks and inserted into the database, at which point the new data is automatically displayed on the bNAber web application.
bNAber provides multiple menu driven interfaces and an option to submit user edited SQL queries (in development). However, we anticipate that many users will have limited experience in using on-line databases and in this tab we provide access to several queries defined by specific user questions. We have pooled the beta testers of the bNAber database about questions they would like to have answered and clicking on a specific question on the list brings the users to a dedicated query designed to answer this question. bNABer team invites users to provide us with additional questions, for which new custom interfaces will be made available.
Access to general help, including links to several precalculated examples, database tutorial and on-line supplemental material are provided in this tab. Additionally, options to export the bNAbs database is available here. Options to download specific tables and data for subsets of bNAbs are also available for other pages.
Use of the bNAber website and database has been optimized for the latest version of Mozilla Firefox, which is open source and available free for download at www.mozilla.org. Regardless of which web browser is used, cookies must be enabled for the website to function properly. bNAber is best viewed with a desktop resolution width greater than 1048px.
Some tools, such as Structure Viewer and POSA require the latest version of java, which is available at http://java.com/en/download/index.jsp .
Currently there are known compatibility issues with Java and Safari, and we recommend using Firefox as a workaround in the interim. Tablets and mobile devices tested were able to access bNAber as expected, but many mobile web browsers do not support Java, and will not be able to view POSA or Structure Viewer.
Any questions regarding the technical requirements of the website, or if you wish to report a suspected technical limitation not listed here, should be addressed to firstname.lastname@example.orgPOSA Home | Other Servers | Godzik Lab | Sanford-Burnham for Medical Research Institute | POSA Contact | bNAber Contact