BIOPYDB: A Dynamic Human Cell Specific Biochemical Pathway Database with Advanced Computational Analyses Platform

3.1 Pathway Nomenclature and Ontology

A standard nomenclature system with structured vocabulary is currently very much required for indexing different biochemical pathways in the respective databases [14]. It is observed that although the biochemical pathway databases have made a significant progress of archiving the pathway data in various modes and file formats, but they are not yet concordant with a specific pathway nomenclature system [14]. The naming convention of the signaling pathways used by the databases most often depend on the names of the ligands or receptor molecules (e.g. Hedgehog pathway or Notch pathway), and some cases it is based on the main target transcription factor of the pathway (e.g. NF-κB signaling pathway). The ambiguity of naming the pathways is also a common factor of the databases, in which a particular pathway is named in multiple ways and then indexed into the database [14]. Hence, by following a consensus naming system with well defined vocabulary and a hierarchical tree of the pathway based ontology will be of great importance to the pathway data curators for indexing the pathways systematically and respectively searching, analyzing the pathway data more smoothly. The ontology tree defined by BioPortal for annotating the rat, mouse and human genes into pathway terms would be most appropriate [15].

Currently, the curators of BIOPYDB are using the similar pathway ontology tree defined by this well accepted and popular portal of biological ontology to index the signaling and disease related pathways into its database. However, to resolve the issues related to the structured vocabularies of naming the pathways, BIOPYDB introduces its own syntax and vocabulary. Currently, BIOPYDB provides 3 different functional (developmental, immunological and cell proliferation pathways) and 1 biological context dependent (disease pathways) nodes of signaling pathways, and has placed them at the top of the hierarchy in the BIOPYDB pathway ontology tree. These top most level nodes are further linked with the child nodes in its down-stream branches. For example, under the node of immune signaling pathways, the cytokine signaling pathways constitute a separate sub-category (child node), which is further classified into distinct interleukin families like IL-2 family, IL-10 family, etc. Each of these interleukin families is further assigned to different interleukin signaling cascades, which are having similar functionalities. More number of different categories will be included in the subsequent updates of this database. To assign the names of the new pathways and make them distinct from each other within a same family and sub-family, the pathways are named according to either ligand (e.g. IL2, IL4) or the main receptor protein (e.g. TLR, T-cell receptor or TCR) molecule, which triggers that particular signaling cascade. The following vocabularies (a, b, c) are used for the nomenclature of the signaling pathways involved in various cellular functions (a, b) and disease pathogenesis (c):

1. X Ligand(s) Stimulated Signaling Pathway(s)

2. Y Receptor(s) Mediated Signaling Pathway(s)

3. Deregulated Signaling Pathways in Z

where, X, Y, and Z represent the name of the Ligand or the family of the Ligands, Single or a family of Receptors molecule, and name of the Disease respectively.

If there are multiple ligands/receptors associated with the signaling pathway, then the logical operator “AND” is used in the vocabularies (“a” and “b”) to separately mention the names of each of the ligands/receptor molecules. Introduction of such detail classification and nomenclature system is useful to the data curators as well as the users for searching/browsing a specific class/family of pathway(s) in the database. It is also helpful to understand and compare the different pathways activated/stimulated under different extra-cellular stimuli. The information about the context specific pathways (e.g. the disease pathways) is also indexed here by using a controlled vocabulary (c), which could be further used to differentiate the deregulated or malfunctioned pathways from its normal counterparts and their involvements into a particular disease (see Supplementary File 1). A flowchart describing the entire process of implementing and processing the pathway nomenclature system in the BIOPYDB database is depicted in Figure 1.

Figure 1:

Flowchart describing the pathway nomenclature system used by the curators of BIOPYDB database.

3.2 Pathway Data Curation and Processing

Each of the interactions/connections/molecular reactions represented in the pathway data/image is acquired from published articles (around 520 articles are manually searched/referred for cross validation of interactions) to provide the appropriate experimental references. One of the main objectives of developing this database is to make a resource of biochemical pathways by feeding minimal information. Hence, to populate the database by feeding optimal amount of data, at first the pathway curators of BIOPYDB have consulted various similar resources (see Table S1 of Supplementary File 1) to gain the basic topology of each of the signaling networks. Most often, the topology and the pathway related data (i.e. number of molecules, interactions etc.) found in the external resources are highly heterogeneous (in terms of pathway names, sub-cellular locations of the pathway species, total number of interacting pathway components etc.) and hence it is required to normalize these piecewise information/data followed by its validation through literature reference, and/or cross-referencing with Protein-Protein Interactions (PPI) or Disease specific (e.g. cancer) databases [5], [14]. Figure 2 represents a schematic diagram of the different resources of pathway data that are used by BIOPYDB pathway data curators to get the information about the basic topology or the core functioning module of the pathway of interest. Later, the data collated from these resources are expanded and reconstructed by including more number of pathway species (inorganic or organic molecules, proteins, lipid, carbohydrate, ions etc.), cross talks, chemical or physical properties of the interactions, etc., which are found to be involved in that particular pathway, as recorded in the literature references, but are not included in any existing pathway databases. The pathways become more up to date, authentic and accurate for the users after inclusion of such comprehensive information from different literature sources.

Figure 2:

External data sources used to collect pathway information for BIOPYDB

In order to build or update the pathways, BIOPYDB uses a specific protocol (see Supplementary File 1) for further pathway data normalization and updating with the latest information and new experimental findings into its database. The following subsections provide a brief description of the protocol used for preparing the BIOPYDB database.

3.3 Processing of Molecules Table

Followed by the nomenclature and ontological assignment, the core molecules (i.e. proteins, genes, metabolites, RNA etc.) involve in a particular pathway of interest are searched in literatures and other databases. Each molecule is further categorized according to its molecular and chemical properties (e.g. Protein, carbohydrate, Gene, Ion, RNA etc.), and is then indexed in the database by its corresponding BIOPYDB Pathway ID. An abbreviated name, which is used in the published literatures or UNIPROT database, is assigned to each molecule. Also, the information of the sub-cellular location of the molecule is extracted from the published literatures and other databases. Moreover, it should be noted that, if multiple sub-cellular locations (e.g. cytoplasm and nucleus) are found to be associated with the same pathway molecule, prefix “Nuc_” is used to annotate the nuclear location to distinguish the pathway species with its cytoplasmic counterpart. No separate prefix is used for cytoplasmic species. For example, GLI1 (an important protein in Hedgehog pathway) is found in the cytoplasm as inactive at the time of no stimuli from hedgehog ligands but is found to be active and translocate into the nucleus followed by the activation of hedgehog signaling event. In this case, GLI1 protein in Hedgehog pathway is named as GLI1 and NUC_GLI1 for its cytoplasmic and nuclear counterparts respectively.

Furthermore, to show the molecular complexes formed by proteins and other molecules in the pathway, the following vocabulary is used. For example, let us consider the following reactions (i.e. E1; E2; E3) from a signaling event in which the complexes are formed as products.

Different types of complexes formed in the signaling cascades by various chemical reactions are indexed in the molecule list by using the vocabularies A:B; A:B:C, and M:M etc.

3.4 Hyper-linking and Annotation of Pathway Molecules

Hyper-linking and annotating of each protein molecule for all the pathways in a database is a time-consuming process, as these require more manual interventions and efforts. In BIOPYDB, minimal pathway information is required to provide from the pathway curator’s end, and its in-built pathway update engine running at the backend automatically processes the hyperlinks of each of the protein molecule with other popular databases. For example, to get the information of the corresponding gene and amino acid sequences of a particular protein, BIOPYDB automatically hyper-links the protein molecule with NCBI-gene and UNIPROT databases respectively. Other popular databases viz. KEGG [16], WIKIPATHWAYS [17] and HUMANCYC [18] are also hyper-linked with the protein molecules to make this database more useful and authentic. To get the information of other interacting protein molecules against a particular protein of interest, each protein molecule of a pathway is provided with the interactive access of protein-protein interaction databases viz. PIPs [19] and STRING [20] databases. Also, to search the diseases associated with each of the protein molecules in a pathway, BIOPYDB provides the hyperlink with GeneCards [21] database, which is a very useful database for acquiring information of protein and disease relationships. The tissue-specific expression of a gene/protein is also an useful information, which is also served in BIOPYDB by providing the hyper-links of the gene/protein molecules with a popular database: TiGER [22] for this purpose. These automated processes of cross-linking the protein molecules with other relevant and popular databases are based on its in-build pathway update engine written in PHP, Perl, and Python languages.

3.5 Processing of Molecular Interactions Table

The molecular interactions or reactions present in each pathway in BIOPYDB are manually curated from the experimental data available in literatures and are successively stored in a master table called “Interactions”. To include a new interaction in this table, the evidence of the interaction should be supported by at least one experimental reference from a published article. Any new interaction, which qualifies this criterion, is then included in the “Interactions” table with its corresponding BIOPYDB pathway ID. The corresponding literature reference of each interaction is then hyper-linked with the associated Pubmed ID of the article. The molecular and chemical properties (e.g. phosphorylation, ubiquitination, dimerization etc.) of different types of biochemical reactions are tabulated with each of the interaction in the “Interactions” table. The sub-cellular location, (e.g. extracellular space, membrane, cytoplasm etc.) in which the interactions are found to occur, is also tabulated with each interaction in this table. To maintain the direction of the reaction cascades, each of the reactions is considered as the directed edges connecting a source molecule with another target molecule (i.e. protein, complex, ions etc.). The source molecule (represented by “Protein A”) is the upstream molecule of the target molecule (represented by Protein B) in the “Interactions” table. All such molecular interactions of a pathway are automatically hyper-linked with iRefIndex database [23], which provides a standardized indexing of the non-redundant molecular interactions by taking the popular protein-protein and molecular interactions databases such as BioGRID [24], IntAct [25], HPRD [26] etc., into consideration. Mapping the interactions with the iRefIndex database will be very much useful for the users to verify as well as collate the information of that particular interaction available in the other popular PPI databases.

3.6 Processing of Disease, Reference and Mathematical Model Tables

Deregulations in the different components of the biochemical pathways and their correlation with several disease pathogenesis are now one of the major research interests to the clinical and experimental biologists. To get such information flawlessly and with less effort, BIOPYDB is providing an interface to search and fetch the protein-disease mapping data from MalaCards: Human Disease Database [27]. The disease data is fetched from this database for each protein mapped with the particular BIOPYDB pathway ID. The diseases mapped with each protein are then tabulated in BIOPYDB web interface and a network picture of protein-disease map is provided for better visualization and further analyses. It should be noted that the protein-disease data provided in BIOPYDB is solely owned by the MalaCards database, and BIOPYDB is providing an advanced, automated interface to fetch and process the data in a more user-friendly process.

BIOPYDB also maintains a dedicated database of all the relevant references or published literatures related to each pathway in “Reference” table. The reference table is generated automatically by fetching the citation information from Pubmed database. The users can get this entire “Reference” list by accessing the pathway-browsing interface of BIOPYDB. Thus the literature mining for getting the information of a particular pathway will be much easier to the experimental biologists. Furthermore, to ease the efforts of theoretical biologists for obtaining information of previously published mathematical models developed for different biochemical pathways, BIOPYDB has also made a database of the literature references related to the mathematical models (e.g. Graph theoretical, Boolean, Ordinary or Partial Differential equations etc.) and tabulated in the “Model” table. The Pubmed id or DOI is also provided for each model and their hyper-links are automatically generated.

3.7 Processing of Pathway Image and Textual Data

There are various file formats used in BIOPYDB for sharing and distributing the pathway data to the users. Users can download the data without any restrictions and charges. For commercial uses, the users are required to take appropriate license from the BIOPYDB developers. The entire process of pathway data processing is described in Figure 3 for better clarification. The pathway data provided by this database is broadly categorized into two categories: (i) “Pathway Image” and (ii) “Textual Data”. Both are automatically processed from the BIOPYDB main data sources (i.e. Molecules list, Interactions list etc.) for further sharing and the downloading purposes. The pathway images are processed in JPEG, SVG, and PNG formats. Two types of pathway images are produced viz. Structured or Hierarchical and Unstructured or Network. The Structured pathway image is produced by allocating the pathway species in a hierarchical fashion, in which the ligands and receptor molecules (present in the extracellular and membrane regions) of a pathway are placed at the left (in left to right hierarchy) or top (in top to bottom hierarchy) in the pathway image. The pathway components found in the cytoplasm and in nucleus are placed accordingly in the subsequent hierarchical levels (i.e. Extracellular region, Membrane, Cytoplasm, Nucleus) respectively. Another hierarchical location “Output” is introduced in the pathway image to signify the target genes/proteins produced at the end of the signaling cascades. The pathway molecules belong to the hierarchical location “Output” are placed at the rightmost or at the bottom of the Structured or Hierarchical image. Though this region does not indicate any physical sub-cellular location, but to distinguish the pathway molecules from the target proteins, inclusion of such location in the pathway image will be very much useful for better simplification. The sub-cellular locations, required to maintain this hierarchy, are fetched from the “Molecules” table generated at the beginning of the pathway curation. On the other hand, the connections/edges between two pathway molecules are fetched from the “Interactions” list. Different color codes and shapes are used to classify different types of molecular interactions enlisted in this table. Thus, an in-silico structured pathway image, resembling the internal cellular environment is reproduced automatically in the database interface. The unstructured pathway image (i.e. network) is also generated to analyze the topological properties of the biochemical signaling pathways. Such type of network is viewed as a directed graph in which the direction of the interaction between the two species of a pathway (i.e. Source and Target) is generated from the information provided in the “Interactions” table. Both of these pathway images (i.e. hierarchical and network) are generated automatically by using the popular graph drawing software: Graphviz [28] and the in-house codes written in PHP.

Figure 3:

Flowchart depicting the process of pathway image construction and textual data preparation in the back-end of database server.

On the other hand, the textual data (i.e. information of molecule, interactions, diseases etc.) of biochemical pathways are generated in BIOPYDB through an automated fashion by fetching the raw data from the database tables’ viz. “Molecules”, “Interactions”, and “Diseases”. The textual data for each pathway are processed in simple flat files i.e. either in TXT or CSV formats or in XML based SBML (Systems Biology Mark-up Language) and BIOPAX (Biological Pathway Exchange) formats [29], [30]. SBML and BIOPAX files are computer-readable, vocabulary based, widely used standard pathway sharing file formats, which are useful for pathway visualization and performing mathematical modeling. The dynamic engine running on the backend of BIOPYDB web services automatically processes these verities types of textual data. To generate runtime BIOPAX format of the biochemical pathways, BIOPYDB uses the open source, third-party tool Sig2BioPAX for converting the biochemical reactions into BioPAX level 3, OWL format [31]. Using such automated process of generating pathway images and textual data, the data in BIOPYDB can be readily updated and thus does not depend on any manual intervention by the database administrators and curators.

3.8 Data Storage System

Relational Database Management System (RDBMS) is used in BIOPYDB for managing its entire database. MySQL database server is used here for this purpose from which the stored data can be accessed through Structured Query Language (SQL). Different types of pathway related data (i.e. molecules, interactions, diseases etc.) are stored in different “Tables”. These tables are the logical objects of the relational database within which the relations are established through unique pathway ids or primary key [32]. The structure of the database is made very simple and can be easily understood. BIOPYDB stores the pathway specific data (within tables) in such a way, so that after inserting or updating the pathway data, the successive outputs of pathway images, annotation and hyper-linking of pathway molecules with other databases, mapping the molecular interactions with iRefIndex database, or generating the textual data for pathway data sharing etc. will be automatically mapped by the corresponding unique pathway IDs and successively stored in the respective tables of the database. This automated process and the structure-function relationship of BIOPYDB are presented more precisely in Figure 4. In this figure, the entire data structure and its relationship with the available features are presented in a simple relational schematic diagram. To understand the database architecture and its interaction procedures with the end-users, a detail description of this diagram is provided in the Supplementary File 1.

Figure 4:

Schematic diagram of BIOPYDB database architecture.

In the backend of this database, it contains the logical objects or “Tables”, which individually store the pathway related data such as Pathway Name, Pathway Info, Overview of each of the pathway, the cross-talks of each pathway, Molecules, Interactions, Diseases, information about the developed mathematical models, References etc. All the tables contain a primary key (Pathway ID), which is used to connect each of the tables while performing the data searching and entry operations in the database. Hence, each and every data in the database stored in the database tables (pathway molecules, interactions, references etc.) are associated with such unique pathway id, which makes the different operations (e.g. insertion, deletion, modification etc.) in the database more accurately using structured query language (SQL) with the help of any dynamic programming language such as PHP. This schema provides this database more fluidity to perform post-processing tasks in an automated fashion by consuming less time and manual interventions.

4.1 Resource of Biochemical Pathways

BIOPYDB is currently providing the biochemical, intra-cellular signaling pathway data under four different categories viz. (i) Developmental Pathways, (ii) Immunological Pathways, (iii) Cell Proliferation Pathways, and (iv) Disease Pathway. In the Developmental Pathways category, two important developmental pathways viz. Hedgehog and Notch pathways are included. Hedgehog pathway is further sub-categorized into three different pathways based on the three homologues of Hedgehog ligands (i.e. Sonic, Desert and Indian Hedgehog). Notch Pathway is also divided into two pathways which are activated by either JAGGED or DELTA ligands. Moreover, BIOPYDB has a rich collection of cell growth regulator pathways eminently represented by the Epidermal Growth Factor Receptor (EGFR/ErbB) mediated Mitogen Activated Protein Kinase (MAPK) pathways under the “Cell Proliferation Pathways” category. The EGFR/ErbB mediated signaling system forms a family of pathways, which is classified here on the basis of ligand-receptor interaction. Apart from the known Growth factors, like the Epidermal Growth Factor (EGF), heparin-binding EGF (HB-EGF) and Neuregulins (NRG1, NRG2, NRG3, and NRG4), a few of the lesser known ligands of the family like Amphiregulin, Betacellulin, Epiregulin are also included. Strikingly, Transforming Growth Factor (TGF-α), a TGF family ligand, due to its structural similarity with the growth factors of EGFR/ErbB family, is classified into the EGFR/ErBb family of signaling. A unique ligand of neuregulin subfamily, Neu Differentitaion Factor (NDF-β), is also included in this category. In contrast to the other NDF isoforms, which generally trans-activate ErbB-3 co-expressed with ErbB-2, NDF-β can also trans-activate ErbB-3 co-expressed with ErbB-1. BIOPYDB also contains an exhaustive list of “Immunological Pathways” that includes pathways related to both innate (e.g. TLR pathways) and adaptive immunity (e.g. T cell signaling pathways). It contains a repertoire of cytokine signaling pathways as well, that includes different interleukins, interferons, tumor necrosis factors, tumor growth factors and colony stimulating factors, all of which plays a pivotal role in regulating the immune-regulatory network. To provide the context-specific pathways, BIOPYDB also provides the signaling pathways, which are found to be deregulated in a particular disease scenario under “Disease Pathways” category. The malfunctioned signaling pathways, which are only stimulated by PDGF, IGF1 and EGF ligands in Glioma, are currently provided in this section. Later, it also intends to provide all other possible pathways, which are responsible for the development of Glioma in human. All these categories of pathways are easily accessible from the “Browse Pathways” tab provided in the home page of BIOPYDB.

To access the information of the pathway of interest, the users are required to click on the corresponding link of the pathways in the pathway browse section. A small description of each pathway, including the pathway statistics (number of molecules, interactions, articles, diseases etc.), will be available in the “Overview” section of the pathway data. Besides, the phenotypic expressions (e.g. cell division, cell growth, apoptosis etc.) and cross-talks with other pathways with proper literature references will be also available in this section for each pathway.

Presently, BIOPYDB contains 3189 molecules (Proteins, Protein complexes, Inorganic molecules, Mutated proteins, Secondary messengers, Phospholipids and Lipid molecules) and 5742 molecular interactions or connections of 46 manually curated pathways. For accuracy and authenticity of the pathway data, each and every interaction included in the pathway is cited (by a published article with its corresponding Pubmed ID). Users can cross verify the pathway data immediately by accessing the hyper-links provided in the “Interactions” tab under the “Browse Pathway” option. It also facilitates to cross-check the pathway data in similar 17 different databases by automatically searching the links in these databases and thus helps to connect the users with these resources immediately.

4.2 Resource of Proteins/Genes Involved in Biochemical Pathways

BIOPYDB can also be used as a resource of signaling protein(s)/gene(s), which is evidenced in the literature regarding direct or indirect influence in the intracellular signal transduction cascades. Currently, it contains 2748 such proteins/genes, which are easily accessible or can be searched through the general “Search” as well as “Advanced Search” option provided in “Home” page. The query result gives a short description of the protein, as well as its genetic and amino acid sequence data from NCBI-GENE and UNIPROT respectively [33], [34]. Users can also get the links of the proteins from other pathway resources like KEGG, WIKIPATHWAYS and HUMANCYC databases [16], [17], [18]. Besides, it provides the hyper-links of the protein-protein interaction databases from the databases PIP; Disease related data from GENECARD and tissue-specific expression pattern data from TiGER database [13], [19], [22].

4.3 Resource of Protein-Protein Interaction Data

Protein-protein interaction data with network representation and proper references are also available through the “Find Interaction” application, available in “Tools” section. The type of interactions (i.e. chemical nature) of each interaction is also available with each of the interaction. Currently, BIOPYDB has included 25 different types of reactions/ interactions/ connections in the signaling networks. The types of interactions are Cholesterol Modification, Physical interaction, Inhibition, Phosphorylation, Activation, Nuclear Translocation, Transcriptional Co-repression, Protein Production, Transcriptional Co-activation, Protein Recruitment, Auto-phosphorylation, Ubiquitination, Transcriptional activation, Complex formation, Stimulation, Dissociation, Dephosphorylation, Homodimerization, Heterodimerization, Deubiquitylation, Calcium exchange, Acetylation, Phospholypase reaction, Proteolytic cleavage, and Enzymatic Reaction [35], [36], [37], [38], [39], [40], [41], [42]. Information related to the type of chemical reactions or interactions occurring between two pathway molecules present in a pathway will help the users of BIOPYDB to understand the biochemical processes, which are regulating the signaling cascades inside the cell. Moreover, to further cross verify the interaction data, each interaction in the database are mapped and hyper-linked with the iRefIndex database [23], which consists non-redundant molecular interaction data collated from different protein-protein interaction databases.

4.4 Resource for Visualizing the Protein-Disease Mapping

It is well known that certain malfunctions caused by the mutation of proteins in biochemical pathways could be responsible for various types of diseases in human body [43], [44], [45]. Therefore, it is worth to mention that protein-disease mapping of the protein molecules in the signaling pathways is also important to understand the importance of that pathway in disease pathology. To include this feature into the database, BIOPYDB provides an interactive interface to dynamically fetch the information of various diseases from protein-disease mapping database: MalaCards [27] and displays the results in a more user-friendly way. All the information provided in this section is solely owned by MalaCards database and the developers of BIOPYDB do not hold any claim related to the protein-disease mapping data. It simply provides a resource for fetching and visualizing protein-disease network as image and tabular formats in the BIOPYDB interface. The diseases associated with each protein of the pathway of interest are automatically hyper-linked with the MalaCards database ID. Currently, BIOPYDB has dynamically mapped around 6897 diseases with the proteins present in the pathways in BIOPYDB database. The disease related information is available through the “Disease” tab under the “Browse Pathway” option. It is also available through the “Find Disease” application available in the “Tools” Section.

4.5 Computational Platform for Pathway Data Analyses

One of the unique features of BIOPYDB is its in-built pathway data analyses platform for performing network (using graph theoretic analysis), logical (using discrete time, semi-dynamic Boolean equations) and dynamic (using ordinary differential equations or ODEs) analyses of biochemical pathways [46], [47], [48]. No other similar resources/databases provide this wide ranges of computational tools all together in a single platform to the common users. Moreover, to make this database more interactive and user-friendly for the large community of experimental and theoretical biologists, developers of BIOPYDB have included various technical features and applications to search, retrieve, annotate or analyse the data stored in the database. The entire resource is easily searchable and does not require any separate hardware or software installation in the user’s local machine. To view the pathway image and network, users are required to use an SVG compatible web browser, which is now available in any modern web browsers, like Internet Explorer (version >9), Mozilla, and Google Chrome etc. A schematic diagram showcasing all the available tools and the stepwise guidelines of their operational protocol is presented in Figure 5. An additional data file containing a brief description of all the technical features available in BIOPYDB is provided in Table S2 of “Supplementary File 1”. Following is the brief descriptions of the technical or computational features available in BIOPYDB.

Figure 5:

Schematic diagram showcasing all the available tools and their working protocols.

4.5.10 Web Interface

The web interface of BIOPYDB is designed in such a way so that users can easily interact and navigate through the database web pages. The “Homepage" of the database contains all the necessary tabs like “About BIOPYDB”, “Browse pathways”, “Downloads”, “Tools”, “Upload Pathway”, “FAQ” and “Contact Us”. The database homepage also provides text “Search” and “Advanced search” options. “Search” option has “auto-suggestion” option enabled, which facilitates more convenient searching of proteins, pathways, and diseases. The left panel of homepage contains “News & Updates” to highlight the subsequent updates and the right panel contains “Current Database Statistics” which is automatically updated if there is any changes/insertion or deletion in the database. In Figure 6, an HTML web page dumped picture is provided to understand the navigation of the web pages from the BIOPYDB home page.

Figure 6:

The web-interface of BIOPYDB.