Introduction

MeDBA aims to provide an integrated, distinctive, practicable, user-friendly platform for metalloenzyme-focused data bank and analysis. It integrates information regarding metalloenzymes, ligands, metalloenzyme-ligand associations with metalloenzyme-specific analysis tools, which will help understand and investigate metalloenzyme-involved contents, and will promote and bridge metalloenzyme-related areas (e.g., biological chemistry, structural and chemical biology, synthetic biology, bioinorganic chemistry, medicinal chemistry, computational chemistry and drug discovery). MeDBA is developed and maintained by Prof. Guo-Bo Li's group.

Metalloenzymes

Metalloenzyme Statistics

The definitions of metalloenzyme categories and overall statistics of the metalloenzymes are summarized in this page.

Metalloenzyme Categories. The metalloenzymes are manually curated to three categories: M-I, M-II, and M-III (as illustrated in Figure 1.1), where
M-I enzyme metal ion(s) are directly involved in catalysis and tightly bound in active site,
M-II enzyme metal ion(s) are directly involved in catalysis but not tightly bound in active site,
and M-III enzyme metal ion(s) are important for enzyme activity (e.g., by improving structural stability) but not present in active site.

Figure 1.1 Illustration of 3 metalloenzyme categories.

Enzyme Class - Metalloenzyme Distribution. The distribution of metalloenzymes according to metalloenzyme category and enzyme classification is displayed in the form of doughnut charts (Figure 1.2).

Figure 1.2 Illustration of “Enzyme Class - Metalloenzyme Distribution”.

Metal Ion - Metalloenzyme Distribution. The distribution of metalloenzymes according to metal ions is illustrated in Figure 1.3, where each bar represents the number of metalloenzymes that contain the corresponding metal ion.

Figure 1.3 Illustration of “Metal Ion - Metalloenzyme Distribution”.

Metalloenzyme - ICD-11 Diseases (Top20). The distribution of metalloenzymes in various diseases (showing the top-20 diseases in ICD-11 disease taxonomy) is illustrated in Figure 1.4.

Figure 1.4 Illustration of “Metalloenzyme - ICD-11 Diseases (Top20)”.

Metalloenzyme - MeSH Diseases (Top20). The distribution of metalloenzymes in various diseases (showing the top-20 diseases in MeSH disease taxonomy) is illustrated in Fig1.5.

Figure 1.5 Illustration of “Metalloenzyme - MeSH Diseases (Top20)”.

Metalloenzyme Search

On the “Metalloenzyme Search” page, metalloenzymes and related general information can be obtained as follows (shown in Figure 1.6):
① Select a search field (as shown in the yellow box of Figure 1.6) from metalloenzyme, metal ion, UniProt ID, gene name, organism, PDB code, EC number, UniProt keywords, metal site class, and type in search keywords.
② Click the “SEARCH” button to search.
③ Filter the search results with options including “Metalloenzyme Category”, “Enzyme Class”, “Metal Ions/Metal-Containing Cofactor”.
④ Download the search results on current page.

Figure 1.6 Illustration of the “Metalloenzyme Search” page.

Notably, the search field 'metalloenzyme' supports recommended name and alternative names from UniProt. It is also worth noting that plenty of internal links are provided to navigate to various detail pages. For example, clicking the UniProt ID will lead users to the metalloenzyme detail page, where further information of a specific metalloenzyme will be available. Links to the PDB detail page (clicking the PDB code) and the metal site class detail page (clicking the metal site class ID) allow users to get a more comprehensive view of specific PDB structure and metal site information, respectively.
Users can also get access to the original data resources ( e.g., UniProt, InterPro, PDB databases) by clicking on the blue arrows.

Metalloenzyme Families

The metalloenzyme families are exhibited in a hierarchical tree structure according to InterPro protein families, facilitating the exploration of metalloenzyme family distribution. On the “Metalloenzyme Families” page (as shown in Figure 1.7), detailed description about a desired family and relevant metalloenzymes can be found as follows:
① Type in the name (e.g., “carbonic anhydrase”) or InterPro ID (e.g., “IPR001765”) of a specific metalloenzyme family.
② Click the submit button to search the related families which will be highlighted in red.
③ Click the interested metalloenzyme family to investigate its detail description and related metalloenzymes (as illustrated in the yellow box).

Figure 1.7 Illustration of the “Metalloenzyme Families” page.

In the yellow box, family structure, description, function (e.g., molecular function, biological process) and related metalloenzymes information (e.g., UniProt ID, category, name, organism) are available which will be helpful for investigating biofunctions or catalytic mechanisms particularly for uncharacterized metalloenzymes.

Metalloenzyme Diseases

On the “Metalloenzyme Diseases” page, similar with the “Metalloenzyme Families” page, metalloenzyme-related diseases are exhibited in ICD-11 or MeSH disease taxonomy. To search for expected diseases, users can follow the simple steps (Figure 1.8):
① Select a disease taxonomy out of “ICD-11” and “MeSH” (“MeSH” for example).
② Type in a disease name (e.g., “Andersen Syndrome”) or disease identifier (MeSH Disease ID: “D050030”).
③ Click the submit button to search for the related diseases.
④ Click the interested disease to see the detail in the right yellow box.

Figure 1.8 Illustration of the “Metalloenzyme Diseases” page.

The detail information of the diseases comprises of disease attribution, introduction and related metalloenzymes along with UniProt ID, name and corresponding clinical status, which will be beneficial for drug discovery particularly targeting multiple metalloenzymes.

Structure Statistics

The “Structure Statistics” module summarizes the information of metalloenzyme structures from various aspects.

Number of Metalloenzyme Structures Since 1980. The line chart in Figure 1.9 reveals the rising trend of metalloenzyme structures since 1980.

Figure 1.9 Illustration of the “Number of Metalloenzyme Structures Since 1980”.

Metal Ion - Metalloenzymes with Structures. The distribution of the metalloenzymes (counted by UniProt ID) with at least one experimental structure according to metalloenzyme category and metal ion is summarized in Figure 1.10.

Figure 1.10 Illustration of the “Metal Ion - Metalloenzymes with Structures”.

Metal Ion - Metalloenzyme Structures. The distribution of the metalloenzymes experimental structures (counted by PDB ID) according to metalloenzyme category and metal ion is summarized in Figure 1.11.

Figure 1.11 Illustration of the “Metal Ion - Metalloenzymes Structures”.

Species (Top 10) - Metalloenzyme Structures. The distribution of metalloenzymes (the inner pie charts) and relevant structures (the outer doughnut charts) according to species (top 10) and metalloenzyme category is shown in the Figure 1.12.

Figure 1.12 Illustration of the “Species (Top 10) - Metalloenzyme Structures”.

Enzyme Class - Metalloenzyme Complex Structures. The distribution of metalloenzyme-relevant complex structures according to enzyme classification and metalloenzyme category summarized in Figure 1.13.

Figure 1.13 Illustration of the “Enzyme Class - Metalloenzyme Complex Structures”.

Structure

Given the unique characteristics of metalloenzymes which contain structurally and/or catalytically important metal ions, metalloenzyme structures are of special concern. On the “Structure” page, all experimental structures and corresponding chains of the identified metalloenzymes are listed which can be queried by the following steps as shown in Figure 1.14:
① Select a search field from PDB code, metalloenzyme, gene name, ligand code, metal-containing cofactor, number of metal ions, metal ions and type in keywords.
② Click the “SEARCH” button to search.
③ Filter the results through metalloenzyme category and enzyme class.
④ Download the search result.

Figure 1.14 Illustration of the “Structure” page and an example of searching "NDM".

Metal Site

We grouped all metal sites extracted by MeCOM according to their metal ions and metal chelates, which resulted in a total of 1738 metal site classes. On this page, users are able to search the metal site classes as follows (see in Figure 1.15):
① Select a search field from UniProt ID, metalloenzyme, PDB code, metal ions, metal number, metal chelates, metal-containing cofactor and type in keywords.
② Click the “SEARCH” button to search.
③ Narrow the search results by metal ions.
④ Download the metal site class information (metal site class ID, number of metal ions, metal ions, metal chelates, metalloenzyme count and corresponding UniProt IDs).

Figure 1.15 Illustration of the “Metal Site” page and an example of searching "NDM".

Structure Superimposition

In the “Structure Superimposition” module, supported by MICAN, users can perform metalloenzyme structure superimposition, visualization, and comparison with simple steps as follows (as shown in Figure 1.16):
① Select two PDB structures with the aid of the search bar which supports query by PDB code, UniProt ID and metalloenzyme name.
② Submit and wait for a few seconds.
③ Analyze the calculated similarity scores and the 3D visualization of the aligned structures (freely downloadable) in the graphical interface (as shown in the yellow box).
To improve users’ experience, several comparison examples and latest job history are displayed in the blue box of Figure 1.16.

Figure 1.16 Workflow of the “Structure Superimposition” module (reference PDB code: 2OVX, query PDB code: 1RM8).

Metal Active Site Comparison

The “Metal Active Site Comparison” module is based on MeCOM, which is developed by our group and tailored for metal site comparison. Likewise, users only need to take very few steps as shown in Figure 1.17 to conduct metal active site comparison:
① Select two PDB structures by searching various fields (including PDB code, active site metal ions, metalloenzyme name, UniProt ID) and choose the superimposition mode (“Metal/Alpha-C Atoms” or “Pharmacophore”).
② Submit and wait.
③ Analyze the similarity scores and aligned structures (in the yellow box), which can be downloaded for local analysis.

Figure 1.17 Workflow of the “Metal Active Site Comparison” module (reference metal site: 4tz9_1, query metal site: 5wih_1).

Similar with the “Structure Superimposition” module, several comparison examples and latest job history are provided as seen in the blue box of Figure 1.17. Notably, users can choose metal/α-carbon pattern or pharmacophore pattern to perform metalloenzyme active site comparison, which might lead to different results. As for the 3D visualization, the surface residues around the metal site are represented as sticks and metals are represented as spheres. Optionally, the cartoon representations of the aligned structures can be switched off to focus on the metal sites. Totally, “Metal Active Site Comparison” module enables users to gain a deeper insight into the similarity of two metalloenzyme metal sites.

DNA/RNA-Binding Metalloenzymes

On the “DNA/RNA-Binding Metalloenzymes” page, users can follow the simple steps (Figure 1.18) to retrieve expected metalloenzymes:
① Select a search field from metalloenzyme, UniProt ID, PDB code, DNA/RNA and input keywords.
② Click the “SEARCH” button to search.
③ Narrow the search results by metalloenzyme category.
④ Download the DNA/RNA binding information.

Figure 1.18 Illustration of the “DNA/RNA-Binding Metalloenzymes” page.

Disordered Metalloenzymes

A list of disordered metalloenzymes, which are either completely disordered or contain large disordered regions in their native state, are independently provided on the “Disordered Metalloenzymes” page. As shown in Figure 1.19, users can search, filter, browse and download corresponding information, especially the disordered rates and disordered sequences.

Figure 1.19 Illustration of the “Disordered Metalloenzymes” page.

Metalloenzyme Detail Page

To inquire the information of a specific metalloenzyme in detail, users will be guided to the metalloenzyme detail page by clicking the UniProt ID on the “Metalloenzyme Search”, “Metalloenzyme Families” and other pages in MeDBA. As shown in Figure 1.20, there is a brief summary panel at the left side of the detail page (in the yellow box), which summarizes the metalloenzyme category, number of related diseases (by ICD_11 and MeSH taxonomy separately), PDB structures and involved ligands, but also provides fast navigations to 4 main sections including “Information”, “Structure”, “Metal Site” and “Ligand”:
① The “Information” section holds the basic information (e.g., metalloenzyme name, gene name, organism, catalytic activity), InterPro family, distribution specificity and related diseases. Notably, there are a number of internal and external links which provide further information. For example, clicking the InterPro family name and the blue arrow behind will respectively lead users to corresponding “Metalloenzyme Families” page and original InterPro website, where overall information about the metalloenzyme family can be obtained.
② In the “Structure” section, protein sequence, protein domain, experimental and predicted 3D structures are available. To visualize the 3D structure, the backbone is represented as cartoon and the metals are shown as spheres. Users can simply switch the 3D structure displayed in the graphical interface by clicking the “3D view” button right behind the structure identifier. Clicking the PDB code will navigate users to the specific PDB detail page .
③ In the “Metal Site” section, the corresponding metal site classes along with metal ions and metal chelates are summarized, which will guide users to the metal site class detail pages for further metal site information.
④ The “Ligand” section provides related ligands along with their ligand IDs, chemical structures, original ligand identifiers, association types and corresponding association contents. Especially, according to data resources, the association types include “Drug”, “Complex Structure”, “Bioactivity”, “Reaction”, “PROTAC”, which can be used as filter options to concentrate on desired ligands. Here, all ligand IDs and chemical structures (SMILES) can be downloaded. Clicking the ligand ID will lead users to sepcific ligand detail page.

Figure 1.20 Overview of the metalloenzyme detail page (UniProt ID: O43570)

PDB Detail Page

Generally, by clicking PDB codes present in MeDBA, users will be guided to the PDB detail page, where there are three main sections including PDB basic information, metalloenzyme information, structure details as shown in Figure 1.21:
① PDB basic information includes title, deposit date, release date, data method, resolution, R-work, R-free, mutation and corresponding reference.
② “Information” section holds the metalloenzyme information including metalloenzyme category, UniProt ID (linked to metalloenzyme detail page), gene name, enzyme class, metalloenzyme family, related chains in PDB structure, sequence length and number of mutations involved.
③ In the “Structure” section, the chemical structures of the ligands and metal-binding pharmacophores (MBP) are displayed. Users can access the ligand detail page and related MBP information by clicking the ligand ID and MBP ID respectively. Notably, the metal site information is summarized, including involved metal site classes, metal ions, binding geometry and details of the metal chelates (residue name, residue ID, chelate atom and corresponding bond length). Moreover, the metal site and ligand of the metalloenzyme structure are visualized in the graphical window, which enables users to investigate the metal coordination features.

Figure 1.21 Overview of PDB detail page (PDB ID: 4KP5)

Metal Site Class Detail Page

Users can get access to the metal site class detail page by clicking the metal site class ID which can be found on the “Metalloenzyme Search” page, “Metal Site” page, etc. As shown in Figure 1.22, this page contains two main sections including “Metal Site Class” and “Related Metalloenzymes”:
① In the “Metal Site Class” section, the 3D visualization of the metal site composition and detailed description about the metal ions and metal chelates are provided.
② In the “Related Metalloenzymes” section, related metalloenzymes, PDB codes and ligands are listed. Clicking the “View” button in the “3D View” column will visualize the corresponding metal site.

Figure 1.22 Overview of metal site class detail page (metal site class: “msc_01589”, metal site: “4kp5_1”)

To facilitate observation of the metal-binding site, multiple display options are designed as shown in the yellow box of Figure 1.22, which enables user to freely select or cancel the representations of “Metal Site”, “Ligand”, “Surface Residues”, “Possible Pockets”, “Cartoon”. Take the metal site “6e6i_1” of “G2IN02” in the detail page of “msc_01589” for example (see in Figure 1.23):
① “Metal Site” and “Cartoon” are selected by default, which displays the chelate residues as sticks (green carbons), metals as spheres and backbone as cartoon.
② If there is a ligand, selecting the “Ligand” option will visualize it as sticks (yellow carbons).
③ Once “Surface Residues” option is selected, the residues surrounding the metal site will be represented as thinner sticks with green carbons.
④ Moreover, possible active pockets will be displayed as a group of light blue dots around the metal site if “Possible Pockets” is selected.
⑤ Users can alternatively switch off the cartoon representation of the backbone by cancelling the “Cartoon” option.
⑥ Users are able to key-in “All” option to turn on all representations.

Figure 1.23 Examples of metal site visualization (metal site class: “msc_01589”, UniProt ID: “G2IN02”, metal site: “6e6i_1”) in metal site class detail page: ① Visualization of “Metal Site” and “Cartoon” options by default. ② Visualization of “Ligand” along with “Metal Site” and “Cartoon” options. ③ Visualization of “Surface Residues” along with “Metal Site”, and “Cartoon” options. ④ Visualization of “Possible Pockets” along with “Metal Site”, “Cartoon” and “Ligand” options. ⑤ Switching off “Cartoon” option to hide the backbone of the metalloenzyme. ⑥ Selecting all representations by “All” option.

Ligands

Ligand Statistics

The “Ligand Statistics” module gives an intuitive view of all the metalloenzyme-related ligands.

General Statistics. The number of bioactive molecules, substrates/products, PROTACs, drugs, drug candidates are summarized for each metalloenzyme category as shown in Figure 2.1.

Figure 2.1 General statistics of metalloenzyme-related ligands.

Drug Target – Metalloenzymes. The distribution of drugs targeting metalloenzymes according to clinical phase (‘Approved’, ‘In clinical trial’, and ‘Under research’) and metalloenzyme category is shown in Figure 2.2.

Figure 2.2 Illustration of “Drug Target – Metalloenzymes”.

Metalloenzyme Drug Target - Enzyme Class. The distribution of metalloenzymes that have drugs or drug candidates according to enzyme classification.

Figure 2.3 Illustration of “Drug Target – Metalloenzymes”.

Metalloenzyme Drug Types. As shown in Figure 2.4, the distribution of drugs according to drug types is summarized for each metalloenzyme category.

Figure 2.4 Illustration of “Metalloenzyme Drug Types”.

Metalloenzyme Drug Indications (Top 10). The statistic of metalloenzyme-relevent drug indications (top-10) is displayed in the form of doughnut charts in Figure 2.5.

Figure 2.5 Illustration of “Metalloenzyme Drug Indications (Top 10)”.

Metalloenzyme – Drugs (Top 20). As shown in Figure 2.6, the number of drugs for top-20 ranking metalloenzymes is summarized.

Figure 2.6 Illustration of “Metalloenzyme – Drugs (Top 20)”.

Ligand Search

All metalloenzyme-related ligands can be retrieved from “Ligand Search” module. Users are able to obtain expected ligands by searching UniProt ID, metalloenzyme name and ligand ID. In particular, substructure search is supported by inputting a SMILES string in the search bar or drawing a molecule structure, which is especially helpful for users to find interested ligands bearing a given substructure or core scaffold in a few seconds as shown in Figure 2.7.
Furthermore, information contents about the ligands are summarized according to the data resources and special properties of them. In other words, a ligand is labelled at least one feature out of “Drug”, “Drug Candidate”, “Ligand has a PDB structure”, “Bioactive Molecule”, “Substrate or Product”, “PROTAC”, “Macrocycle”, “Spiro Compound”, which can be also used as filter options to narrow the results. To learn more about a specific ligand, clicking the ligand ID will navigate to the ligand detail page.

Figure 2.7 The interface of “Ligand Search” module and an example of substructure search by inputting SMILES string “CC(=O)C1CO1” or drawing the molecule structure in the JSME applet window.

Bioactive Molecules

On the “Bioactive Molecules” page, all the collected bioactivity records are provided (Figure 2.8). To focus on the records of a specific metalloenzyme or ligand, users are able to search by metalloenzyme name, UniProt ID or ligand ID. Here, filtering by metalloenzyme category, assay type (IC50, EC50, Ki, Kd) and activity source (BindingDB, ChEMBL) allows users to further exclude unexpected records.
The search results contain the ligand ID, chemical structure, UniProt ID, metalloenzyme name and category, bioactivity, activity source and original reference. To help users learn more about a specific bioactivity record, clicking on the blue arrows behind the activity source and reference will link to the resource databases and original references, respectively.

Figure 2.8 The interface of “Bioactive Molecules” page and the results of searching metalloenzyme with “Carbonic anhydrase” and filtering by assay type “IC50” and activity source “ChEMBL”.

Substrates/Products

In the “Substrates/Products” module, the metalloenzyme-meditated reactions and corresponding substrates/products are provided (Figure 2.9). Users can also search specific fields including metalloenzyme name, UniProt ID and ligand ID, ligand name and Rhea ID to look for reaction records of interest which contain detail information of the reactions (e.g., reaction equation, Rhea ID, ligand ID and chemical structure of substrate/product, etc) and related metalloenzymes with category and UniProt ID. In particular, the graphical reaction equation is designed to illustrate the enzymatic reactions, which can be simply triggered by clicking the “show reaction” button in each record. In addition, there are four reaction directions from Rhea database including “UN”, “LR”, “RL”, “BI”, which refer to undefined, left to right, right to left, bidirectional respectively.

Figure 2.9 The interface of “Substrates/Products” page and the results of searching metalloenzyme with “Histone deacetylase”.

PROTACs

On the “PROTACs” page, users are able to search, filter, browse and download the information of metalloenzyme-related PROTACs (Figure 2.10). Here, metalloenzyme information, ligand ID, PROTAC structure, PROTAC name, E3 ligase, DC50 and source reference are included and freely downloadable.

Figure 2.10 The interface of “PROTACs” page and the related PROTACs of “M-I” metalloenzyme category.

Macrocycles

On the “Macrocycles” page, all collected macrocyclic compounds (ring architectures with 12 or more atoms) are independently provided along with ring information, related metalloenzymes and bioactivity records if available. Especially, the atom number of the largest ring, number of rings, number of rotatable bonds can be obtained (Figure 2.11).

Figure 2.11 The interface of “Macrocycles” page.

Spiro Compounds

On the “Spiro Compounds” page, the number of the spiro atoms, number of rings, number of rotatable bonds, related metalloenzymes and corresponding bioactivity records of the identified spiro compounds are provided (Figure 2.12).

Figure 2.12 The interface of “Spiro Compounds” page and an example of searching “Histon deacetylase”.

Approved Drugs

On the “Approved Drugs” page, the approved drugs targeting metalloenzymes are listed (Figure 2.13).
On the top of the page, “Indications - No. of Drugs” (the number of metalloenzyme-targeting drugs for top 10 indications), “Metalloenzyme Distribution” (distribution of metalloenzymes treated as therapeutic targets), “Drug Distribution” (distribution of drugs targeting metalloenzymes according to metalloenzyme category), “Drug Types” (distribution of drugs according to drug types) are summarized to give an overall view of the collected drugs.
The ligand ID, chemical structure, metalloenzyme target, detailed drug information (including drug type, clinical indication, mechanism of action) are available. What’s more, the metalloenzyme category, drug type, mechanism of action can be used to narrow down the scope of drugs displayed in the table. Notably, clicking a specific clinical indication will navigate to corresponding “Metalloenzyme Diseases” page for detail information.

Figure 2.13 The interface of “Approved Drugs” page and drugs targeting M-I metalloenzymes filtered by metalloenzyme category.

Agents in Clinical Trial

Sharing similar interface and usage with the “Approved Drugs” page, the “Agents in Clinical Trial” page summarizes all the drug candidates targeting metalloenzymes of each clinical phase (Figure 2.14).

Figure 2.14 The interface of “Agents in Clinical Trial” page.

Ligand Detail Page

Clicking the ligand ID (e.g., “MeLig00000311”) allocated by MeDBA will guide users to the ligand detail page which provides overall ligand information. On the ligand detail page, basic information (e.g., information content, InChI, SMILES, molecule weight) and metalloenzyme information (e.g., UniProt ID, metalloenzyme name and category, ligand source, association type, association content) can be obtained (Figure 2.15). To help focus on expected metalloenzymes, users can filter the records by selecting a specific metalloenzyme category or association type. Here, the UniProt ID, metalloenzyme category and name of all related metalloenzymes are free to download.

Figure 2.15 The interface of ligand detail page (ligand ID: “MeLig00000311”).

Associations

Metalloenzyme-Ligand Interactions

On the “Metalloenzyme-Ligand Interactions” page, a list of metalloenzyme-ligand complex structures identified by MeCOM are provided. Users are able to search for expected complex structure by selecting a search field (e.g., PDB code, metalloenzyme name, Unirot ID, ligand code, metal ions), typing in keywords and clicking the search button (Figure 3.1). Particularly, by clicking the “3D view” button in the “PDB” column, the corresponding protein-ligand interactions will be visualized in the right graphical window, where the metalloenzyme residues are represented as sticks with green carbons, the metals are represented as spheres, the ligands are shown as sticks with yellow carbons. Through the 3D visualization, the metal binding modes and ligand interactions of metalloenzymes will be more easily understood.

Figure 3.1 The interface of “Metalloenzyme-Ligand Interactions” page (PDB code: “1A42”).

Metalloenzyme-Protein Interactions

Similarly, the “Metalloenzyme-Protein Interactions” page provides users with the metalloenzyme-protein complexes information including the PDB code, basic information of metalloenzyme as well as the name and UniProt ID of the interacting protein. As for the 3D visualization (Figure 3.2 ①), the PDB structures are entirely represented as cartoons, with the metalloenzyme chains colored green, the interacting protein chains colored blue and other chains colored white.

Figure 3.2 Examples of 3D visualization for metalloenzyme-protein and -DNA/RNA complexes. ① Visualization of histone deacetylase 1(UniProt ID: “Q13547”, green) complexed with metastasis-associated protein MTA1 (UniProt ID: “Q13330”, blue) in “Metalloenzyme-Protein Interaction” module (PDB code: “5ICN”), ② Visualization of histone deacetylase (UniProt ID: “P53973”, green) complexed with DNA (spectrum) in “Metalloenzyme-DNA/RNA Interactions” module (PDB code: “6Z6P”).

Metalloenzyme-DNA/RNA Interactions

The information of metalloenzyme-DNA/RNA complex structures (e.g., PDB code, metalloenzyme name, UniProt ID nucleic acid type and chains) can be obtained in the “Metalloenzyme-DNA/RNA Interactions” module. The complex structures are represented as cartoon in the graphical interface (Figure 3.2 ②), where the metalloenzyme chains, nucleic acid chains and other chains are respectively colored green, “spectrum” (a color scheme in 3DMol), white.

MBP Search

In the “MBP Search” module, a total of 844 excavated MBPs are displayed, which provide not only important links between different metalloenzymes but also clues to establish new metalloenzyme-ligand associations. Users can concentrate on expected MBPs via basic annotations (e.g., PDB code, metalloenzyme name, UniProt ID) and substructure search (Figure 3.3). Furthermore, clicking each MBP structure can get the details of the corresponding metalloenzyme-ligand associations, including the structure view (original ligand structure, metalloenzyme-ligand interaction mode), PDB code, metalloenzyme name, metalloenzyme category, active metal ions and ligand code. Users are able to access metalloenzyme detail page, PDB detail page, ligand detail page for further information by clicking UniProt ID, PDB code, ligand code, respectively.

Figure 3.3 Illustration of the “MBP Search” module and detail metalloenzyme-ligand associations of “MBP0231”.

Profiling by Ligand Similarity

The “Profiling by Ligand Similarity” module provides a web server to predict new potential metalloenzyme-ligand associations on the basis of ligand chemical similarity. Users are able to quickly query similar ligands and corresponding metalloenzymes records (e.g., metalloenzyme categories, bioactivities, complex structures and references) by submitting a molecule by drawing a chemical structure, inputting SMILES or uploading a mol2/sdf file, setting up the profiling parameters including search range (metalloenzyme category subsets and the entire database), similarity approach (FP2, FP3, FP4, MACCS) and similarity cutoff (Figure 3.4). It is worth noting that, if too many similar ligands (>1000) found, top 1000 ligands will be listed and ranked by similarity score. Alternatively, an example molecule, captopril, is available to try the profiling service, which will result in 661 similar ligands and 4022 related metalloenzyme-ligand association records.

Figure 3.4 The interface of “Profiling by Ligand Similarity” module and an example of profiling captopril.