Go back to the MyProteinNet homepage
Our FAQ page
Below you can find: Overview of MyProteinNet, Run MyProteinNet page, and Results
Contact: For questions or comments please email Esti Yeger-Lotem at estiyl@bgu.ac.il
The identification of the molecular pathways active in specific contexts, such as disease states or drug responses, often requires an extensive view of the potential interactions between a subset of proteins. However, this view is not easily obtained: it requires the integration of context-specific protein list or expression data, with up-to-date data of protein-protein interactions (PPIs) that are typically spread across multiple databases. The MyProteinNet web-server allows users to easily create such context-specific PPI networks (interactomes) for organisms and for human tissues. The output is composed of a downloadable, weighted interactome file.
Users of MyProteinNet can integrate their own PPI data with PPI data from 16 major databases and thus construct extensive interactomes for 11 different organisms. By uploading gene lists or data of molecular expression and threshold values users can filter the interactomes to obtain only PPIs between subsets of proteins. MyProteinNet specifically facilitates the construction of human tissue interactomes by providing tissue expression data from three extensive resources (Novartis gene atlas (GNF), Human Protein Atlas (HPA), and Illumina bodyMap2.0). The output interactomes can be uploaded into Cytoscape or used as input for e.g., ResponseNet.
Please note that runs for human and yeast take around 9 minutes, runs for other organisms are typically faster
Run time depends on the number of PPIs available for the organism
Every page contains the following three icons in the top right of the screen:
– Link to MyProteinNet homepage
– Opens MyProteinNet homepage in a new tab
– Link to MyProteinNet tutorial
Left-side menu – links to tutorial, FAQ, output example and contact. Below are additional menu options Run MyProteinNet: Click to submit a new job to MyProteinNet. This is the default page presented.
Load Session: For returning users that logged into MyProteinNet, click to load previous jobs.
Login / Change User: By default users are logged in as guests. To allow access to your jobs at other times or from other computers without saving job links, use this option and login as a user. User jobs are stored for 30 days.
MyProteinNet form:
Organism: We currently support 11 organisms including Homo sapiens (Human), Saccharomyces cerevisiae (budding yeast) and Mus musculus (Mouse).
Select protein interaction databases: All the supported databases allow automatic download through PSICQUIC. The first four databases have been selected by default. Interactions data from each database are automatically updated every two weeks.
Below are the supported databases.
| Database name | Database web address |
| BioGrid | http://thebiogrid.org |
| MINT | http://mint.bio.uniroma2.it/mint/Welcome.do |
| DIP | http://dip.doe-mbi.ucla.edu/dip/Main.cgi |
| IntAct | http://www.ebi.ac.uk/intact/ |
| BIND | http://baderlab.org/BINDTranslation |
| InnateDB | http://www.innatedb.com |
| InteroPorc | http://biodev.extra.cea.fr/interoporc/Default.aspx |
| MPIDB | http://jcvi.org/mpidb/about.php |
| MatrixDB | http://matrixdb.ibcp.fr |
| Reactome-FIs | http://www.reactome.org |
| Reactome | http://www.reactome.org |
| SPIKE | http://www.cs.tau.ac.il/~spike/ |
| STRING | http://string-db.org |
| TopFind | http://clipserve.clip.ubc.ca/topfind |
| Uniprot | http://www.uniprot.org |
| VirHostNet | http://pbildb1.univ-lyon1.fr/virhostnet/login.php |
Add your interactions data:You can add interactions to the set of interactions selected above. Alternatively, you can skip the database download and use your data only. For example, if you want to test several threshold values and save run time, we suggest that you use the unfiltered interactome from the first run, upload this file, specify a different threshold and run.
The line format of the interactions data file is as follows:
source name,target name,source type,target type,interaction type,weight,directionality
For example: ENSG00000116030, ENSG00000008196, Protein, Protein, PPI, 0.080077, 0, 1.0
Source and target names: should be in gene symbol, Ensembl gene identifier or Entrez gene identifier.
Source and target types: types are ‘Protein’, ‘Gene’ and ‘MIR’. You can also add new types. Nodes with similar names but different types will be represented by different nodes in the network.
Interaction type:types are ‘PPI’ for protein-protein interactions, ‘TR’ for transcription regulation interactions, and ‘MIR’ for microRNA-target regulation. You can also add new types.
Weight: value between 0 and 1
Directionality: 0 means undirected interaction, 1 means directed interaction from source to target.
In the end of each run, unidentified source or target node names will appear in the output log.
Unidentified means we could not find this identifier in our downloaded table of identifiers obtained from biomart.
Filter output network:Use this menu to create context-specific networks. Only interactions between selected proteins will be included in the output interactome.
Filter by tissue expression profiles: Only interactions involving genes and proteins with expression level ≥threshold will be included in the output interactome. Human tissue expression profiles were collected from three extensive resources: Illumina bodyMap2.0, Novartis gene atlas (GNF) and the Human Protein Atlas (HPA).
Choose tissue: Allows you to select one of 16 supported tissues. Upon selecting a tissue the form will change to display the available expression profile for that tissue.
Choose expression data for tissue: Allows you to specify the expression resource, which appears as <tissue>@<resource>. You may select more than one resource, in which case MyProteinNet will unite them: a node will be included in the output network if its expression level was ≥threshold in one or more profiles.
References to the expression resources:
GNF: Su AI, Wiltshire T, Batalov S, Lapp H, Ching KA, Block D, Zhang J, Soden R, Hayakawa M, Kreiman G, et al. A gene atlas of the mouse and human protein-encoding transcriptomes. Proc. Natl Acad. Sci. USA.2004;101:6062–6067
HPA: Berglund L, Bjorling E, Oksvold P, Fagerberg L, Asplund A, Szigyarto CA, Persson A, Ottosson J, Wernerus H, Nilsson P, et al. A genecentric Human Protein Atlas for expression profiles based on antibodies. Mol. Cell. Proteomics. 2008;7:2019–2027.
Bodymap2.0: Bradley RK, Merkin J, Lambert NJ, Burge CB. Alternative splicing of RNA triplets is often regulated and accelerates proteome evolution. PLoS Biol. 2012;10:e1001229.
Enter thresholds for databases: Threshold can be any number. Default values are the values we used to calculate the tissue specific interactomes in TissueNet.
Filter by your expression data: Upload expression profile. Expression data line format is: node name,value,p-value
We also accept files containing only name names or node names and expression; however line format should be identical across the file.
Filtering options:
Greater thanlimit to genes with expression value ≥threshold
Lower thanlimit to genes with expression value ≤threshold
Absolute value greater than limit to genes with absolute expression value > threshold
Expression Threshold: Value is any number, such as: 535, -2.5, 1e-1
p-value upper limit:Limit to genes with p-value ≤threshold. Value is y number between 0,1, such as 10e-7, 0.001, 1e-5
Filter by gene ontology (GO) terms: Limit to interactions between proteins annotated with at least one selected GO term. You can insert GO terms, such as GO:0045727, or use Select GO terms menu below, which offers auto completion. Alternatively, you can upload a file with selected GO ids, such as:
GO:0045727
GO:0045631
…
Advanced Parameters
Choose weighting scheme: use this option to add interaction weights to the output interactome. Weights are between 0,1. Uniform weights reflect reliability of detection method(s), while GO process biased also favors selected GO terms. The default option is Uniform weights.
Keywords:Only biological processes containing these keywords will be used to weigh the interactions. For example, upon selecting keyword ‘regulat’ terms containing the word regulation, regulatory, or regulator will be considered. You may choose any combination between the 16 options by using ‘Change’. You may also add your own keywords, separated by commas, in the ‘Other’ text line. The default selection is ‘regulat’, ‘signal’ and ‘respons’.
GO annotation evidence codes:Limit to proteins whose annotation to a GO term is supported by selected evidence codes. You may choose any combination between the 16 options by using ‘Change’. The default selection is ‘EXP’, ‘IDA’, ‘IC’ and ‘TAS’.
GO process size limit:use this option to avoid generalized GO processes. Only processes assigned to a number of genes smaller than the size limit will be considered.
Job details
Job name: Job name will help you find the job at a later time
Job description (optional): A few words or sentences that will appear in the output. Can be used to describe the input parameters.
Run: MyProteinNet will integrate the interactions data from selected sources, filter them and construct a weighted interactome.
Reset:will delete all user input to the form and revert selections to the default options.
Results
Files: Links to output and intermediate interactome files.
Files can be directly uploaded into other network analysis programs, including Cytoscapeand ResponseNet.
Summary: Report of the user input parameters and the numbers of non-redundant nodes and interactions in the output (filtered) and intermediate interactomes.
Output network measurements: allow the user to estimate the connectivity, small world, and scale-free properties of the output interactome.
Average Clustering Coefficient:reflects the connectivity among the interactors of each node.
Number of Connected Components: the number of sub-networks that are not connected to each other.
Size of Largest Connected Component: the number of nodes within the largest connected sub-network.
Degree Distribution:degree is the number of interacting partners per node, distribution shown.
Interaction Weights Distribution:weights range between 0,1, distribution shown.
Log: Report of errors or problems encountered during the run.
Interaction file mapping errors:The identifiers for interactions that were not mapped successfully by MyProteinNet.
Expression file mapping errors: The identifiers for gene names that were not mapped successfully by MyProteinNet.
Non responsive databases:Databases that MyProteinNet could not download data from.
Empty databases for organism:Databases that do not hold data for the selected organism.
Not mapped successfully means we could not find this identifier in our downloaded table of identifiers obtained from biomart.
The job and output will be maintained or 30 days and are accessible through the link at the bottom of the page.