Hint-KB 2
Human INTeractome Knowledge Base v 2.0

Help Page

Table of Contents

  1. Home Page
  2. PPI Search
  3. PPI Complexes
  4. PPI Functions
  5. Datasets
  6. Web API
  7. About
  8. PPI Network Visualization

Home Page

The start page of HINT-KB2. By clicking on the links appearing on the left, you can browse information about Protein-Protein Interactions, Complexes and Functions. Furthermore, you can create your own datasets, read about the Web API, find relative links or contact us. Each one of the tools provided is explained further in the following paragraphs.

In addition, there are some icons that appear in every page. Their functionality is described below:

Positive Interactions Search

This page shows protein-protein interactions that are classified as positive using the 0.075 threshold. The user can specify their own threshold for the classification of the interactions, by changing the value appearing in the threshold value field. After specifying a threshold, the user can see the corresponding positive interactions by clicking on the button. Furthermore, feature normalization can be applied, by checking the option:

Negative Interactions Search

Please check the Positive Interactions Search section above.

Search by Protein

In this page, the user can search for a protein of their choice and see information about its interactions with other proteins, clusters that contain it and functions by which the corresponding protein is characterized. Furthermore the user can see the visualization of the protein interaction network of the specified protein. In order to get help on the visualization part, please check the PPI Network Visualization section.

By clicking on the button, the specified protein's interactions are shown. The results consist of the following tabs:

PPI Complexes

EEMC Clusters

This page contains a list of all the clusters found by applying the EEMC clustering algorithm. The user can see additional information about a cluster, by clicking the button. When clicked, the button redirects the user to the corresponding cluster's page, where the proteins composing the cluster and the function that characterizes the cluster are displayed. The user can return to the cluster list by clicking the corresponding button appearing in the page of each cluster.

Furthermore, the user can download a txt file containing all the clusters, as well as the corresponding proteins and functions for each one. The download button can be found at the bottom of the EEMC Clusters page.

PPI Functions

Search Functions

The user can search a function by entering the corresponding GO Term in the search field. By typing two or more digits, a list of relevant GO Terms appears. When the desired GO Term has been typed, the user can click the button in order to see the results. After the search, three tabs are created:


My Datasets

This page allows an authenticated user to manage their datasets:

The user can View, Download or Delete any of their datasets, by clicking on the corresponding button.

Create a dataset

By using this tool, the user can create their own datasets. They can choose the number of positive and negative protein interactions they want to include in their dataset. In addition, they can apply feature normalization to the results by checking the option:

Users can also choose if they want to include just HPRD positive interactions in the resulting dataset:

Finally, an authenticated user can save their dataset by clicking on the button. The saved datasets are displayed in "My Datasets" page.

For information concerning the resulting data, please refer to the Positive Interactions Search section.


Services Reference

The user can use some of the Representational State Transfer (REST) Services. These services are useful for third party systems that want to integrate the HINT-KB2 functionality.

The user can click on any link at the top of the page in order to see the corresponding service. There is a description for each service, its parameters, the values it returns and examples of using the service.

Web API Client

HINT-KB2 also offers standalone applications so the user can request data. The user can install the application to their computer and use the Web Services the website offers. The steps for installing the applications are described in the page.


This page provides the user with useful links of Human Protein Interaction Databases and databases containing Protein Interaction Annotation Information. By clicking on a link, a new window to the corresponding database will open.

Contact Us

This page provides the user with our contact information. In addition one can send their comments by using the corresponding form.

PPI Network Visualization

General Information

The selected protein is shown in red.

The user can move the nodes around by clicking on them and draging them.

The edge width is based on the confidence score. (greater width means larger score)

Cytoscape Web Controls

Selecting a node/edge

By selecting a node or edge, information will be shown in the Graph Information field. When a node is selected, the protein's Uniprot ID is shown. When an edge is selected, the user can see the interacting proteins' Uniprot IDs and the confidence score of their interaction.

First neighborhood

One can see the first neighborhood of a node by right-clicking on it and choosing "Show First neighbors". The first neighbors and their confidence scores will be shown in the Graph Information field.

Another way of seeing a node's neighbors is by using the Decrease opacity on click option. It can be found at the bottom of the page:

By enabling this option, whenever the user points over a node, every node's opacity is decreased, except for the specific node's neighbors.
Note: Having this option enabled might cause slower Flash response for large networks.

Changing layout

The user can change the graph layout:

Interaction information

By pointing on an edge, a tooltip showing the interacting proteins as well as the confidence score will pop up:

The user can search for a protein or an interaction by entering the Uniprot ID(s) of the protein(s) of their choice.

Second neighborhood

By clicking on the button, the user can see the second neighborhood (2 hops away) of the selected protein. The interactions between the second neighborhood of the selected protein are not shown.

Respectively, when the user is seeing the second neighborhood, they can click on the button, in order to see the first neighborhood of the selected protein.

Developed under the support of Computational Biology and Bioinformatics Group and Pattern Recognition Laboratory of Department of Computer Engineering & Informatics of University of Patras

Human INTeractome Knowledge Base v2.0 2013 - 2019 © Pattern Recognition Laboratory

Creative Commons Licence
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.

Sign In
Fields with * are required.
If you haven't sign in to our tool, we encourage you to complete our registration form.
Create Account