Scope

The LipiDisease tool performs disease enrichment analysis on lipids using biomedical literature data. Kindly note that the resulting associations are only predictions and not the curated data.

Licence

The tool is free for non-commercial and academic use. Kindly contact us for commercial and non-academic use.

Citation

[More, P., Bindila, L., Wild, P., Andrade-Navarro, M. & Fontaine, J.-F. LipiDisease: associate lipids to diseases using literature mining. Bioinformatics btab559 (2021) doi:10.1093/bioinformatics/btab559] (https://academic.oup.com/bioinformatics/advance-article/doi/10.1093/bioinformatics/btab559/6343440)
PMID: 34358314

Data source

The tool uses the data from Lipid Maps®, PubChem, PubMed, and MeSH. The list of lipids and corresponding PubChem CIDs are downloaded from the Lipid Maps. Diseases are identified by MeSH terms from the C branch. Citations associated with diseases are identified using the MeSH annotations in PubMed and citations associated with lipids are identified using the NCBI’s Entrez Programming Utilities.

Using the tool

Selection panel

The selection panel can be found on the left side of the webpage. This is used to sumbit the user input data for specfic analysis.

Analysis type

User can perform following four types of analysis:

1. Lipid-set enrichment
   This perfoms enrichment analysis on a set of lipids, considered altogether. User can define the minimum number of lipids considered for the enrichment using the “Additional options (minimum number of lipids)” settings. There is no limit to the number of lipids the user can enter.
2. Lipids to Diseases
   This associates individual lipids, considered independently, to diseases. There is no limit to the number of lipids the user can enter. Each lipid will by independently associated with the diseases.
3. Dieases to Lipids
   This associates individual diseases, considered independently, to lipids. There is no limit to the number of diseases the user can enter. Each disease will by independently associated with the lipids.
4. Lipid-set enrichment (with fold changes)
   This option performs disease enrichment analysis for a set of lipids, considered altogether, based on the enrichment score (ES). The ES is calculated from the provided lipid-level statistics (e.g. fold changes) using the Fast Gene Set Enrichment Analysis algorithm. Please enter, one per each line, PubChem CIDs for the lipids and corresponding fold changes seperated by comma. The user can define the minimum number of lipids considered for the enrichment using the “Additional options (minimum number of lipids)” settings. There is no limit to the number of lipids the user can enter.

Entering the data

User can enter the data in following two ways:

1. Paste a list of lipids or diseases
   Here user can paste the list of PubChem CIDs for the lipids or MeSH Unique IDs for the diseases.
2. Or select lipids or diseases from the list
   Alternatively, user can select multiple lipids or diseases by selecting their names from the drop down list. The user can start typing the name of desired lipid or disease to filter out the list.

Hint For Lipid-set enrichment and Lipid-set enrichment (with fold changes), at least two lipids are required to perform the analysis
Click on example from the selection panel to load the predefined data

Additional options

1. Minimum number of disease-related citations for a lipid or lipid-related citations for a disease
   Use this option to set minimum number of PubMed articles to be used for defining a reliable lipid-disease association.
2. For a Lipid-set association, minimum number of lipids significantly associated with a disease
   Use this option to define minimum number of lipids required to define the disease enrichment.
3. False Discovery Rate (FDR) cutoff
   Use this option to change the stringency of results.

Click on Submit to perform the analysis!

Output panel

The output panel can be found on the right side of the selection panel. This consists of following four tabs:

1. Enrichment table
   This shows the main analysis result in the form of a table. There will be a message above the table stating if all input terms are matched with the database. If not, missing terms will be displayed and excluded from the analysis.
   The table may be divided in several pages, depending on the length, accessible from a navigation bar at the bottom of the table. By default, the table displays 10 rows per page. This can be configured using the Show NN entries option at the top left corner of the table. The table rows can be filtered for any term from any column using the Search option at the top right corner of the table.
   By default, the table is sorted according to FDR values. The table can be freely sorted by clicking on the column header (Shift+Click for multi-column sorting). Diseases, PubChem CIDs, and citations are linked to MeSH, PubChem, and PubMed respectively. Click on them to open respective links in a new browser tab.

Output table definition

Lipid-set enrichment - Disease: disease associated with specific lipids (links to respective MeSH database entry) - Lipid Count: number of lipids associated with specific lipids - Lipids: PubChem CIDs for the associated lipids (links to respective PubChem entry) - Citations: number of citations used for computing lipid-disease association (links to respective PubMed entry) - Fold change: (number of input lipids significantly associated with the disease in the literature / number of input lipids) / (total number of lipids significantly associated with the disease in the literature / total number of lipids) - p-value: calculated by a Fisher’s exact test - FDR: False Discovery Rate calculateed by Benjamini Hochberg method

Lipids to Diseases - Disease: disease associated with specific lipids (links to related MeSH database entry) - Lipid: name of the lipid associated with specific disease - Lipid_pubchem_cid: PubChem CID for the associated lipid (links to respective PubChem entry) - Citations: number of citations used for computing lipid-disease association (links to respective PubMed entry) - p-value: calculated by a Fisher’s exact test - FDR: False Discovery Rate calculateed by Benjamini Hochberg method

Diseases to Lipids - Lipid: name of the lipid associated with specific disease - Lipid_pubchem_cid - Disease: disease associated with specific lipids (links to related MeSH database entry) - Citations: PubChem CID for the associated lipid (links to respective PubChem entry) - p-value: calculated by a Fisher’s exact test - FDR: False Discovery Rate calculateed by Benjamini Hochberg method

Lipid-set enrichment (with fold changes) - All columns from the Lipid-set enrichment output - ES: Enrichment Score reflecting the overrepresented at the top or bottom of a ranked list (same as Broad GSEA implementation) - NES: Enrichment Score normalized to mean enrichment of random samples of the same size - nMoreExtreme: a number of times a random gene set had a more extreme enrichment scorevalue - leadingEdge: lipids that drive the enrichment

2. Enrichment plot
   A plot, summarizing the output, will be generated in case of Lipid-set association and Lipid-set enrichment. This plot can be locally saved by clicking on the Download the plot as a PDF file button below the plot figure.

3. Download
   The output table can be locally saved by clicking the Download button. This will generate a tab-seperated *.tsv file.

4. Documentation
   This page can be accessed here. Hint For quick help, click on the small help icon to the right of Select analysis type option on the selection panel.

Legal notice

Contact
Email: piyusmor@uni-mainz.de
Telephone: +49-6131-39-23892

Responsible for the content
Dr. Piyush More
Email: piyusmor@uni-mainz.de
Telephone: +49-6131-39-23892
Institute of Organismic and Molecular Evolution (iOME),
Johannes Gutenberg-University Mainz,
Hanns-Dieter-Hüsch-Weg 15, 55128 Mainz
Germany

Privacy and data protection

Extent of validity of these declarations
These declarations cover the web site http://cbdm-01.zdv.uni-mainz.de:3838/piyusmor/LipiDisease/. If you follow links to other external web sites, their use of data may differ.

Cookies
Cookies are small temporary files located on the user’s device accessing Internet used to store specific information. This website does not store any cookies.
You can deactivate the use of cookies in your browser preferences without affecting the performace and usability of the website.

We do not collect user’s any personal data.