UVREK: Development and Analysis of an Expression Profile Knowledgebase of Biomolecules Induced by Ultraviolet Radiation Exposure

Abstract

Humans encounter diverse environmental factors which can have impact on their health. One such environmental factor is ultraviolet (UV) radiation, which is part of the physical component of the exposome. UV radiation is the leading cause of skin cancer and is a significant global health concern. A large body of published research has been conducted to uncover the mechanisms underlying the adverse outcomes of UV radiation exposure on living beings. These studies involve identifying the biomolecules induced upon UV radiation exposure. A few previous efforts have attempted to compile this information in the form of a database, but such earlier efforts have certain limitations. To fill this gap, we present a structured database named UVREK (UltraViolet Radiation Expression Knowledgebase), containing manually curated data on biomolecules induced by UV radiation exposure from the published literature. UVREK has compiled information on 985 genes, 470 proteins, 54 metabolites, and 77 miRNAs along with their metadata. Thereafter, an enrichment analysis performed on the human gene set of the UVREK database showed the importance of transcription-related processes in UV-related response and enrichment of pathways involved in cancer and aging. While significantly contributing toward characterizing the physical component of the exposome, we expect that the compiled data in UVREK will serve as a valuable resource for the development of better UV protection mechanisms such as UV sensors and sunscreens. Noteworthily, UVREK is the only resource to date compiling varied types of biomolecular responses to UV radiation with the corresponding metadata. UVREK is openly accessible at https://cb.imsc.res.in/uvrek/.

Figure 1

Illustrative schematic depicting the connections between data and metadata. This figure showcases the interconnection between various specifications of UV radiation exposure and biomolecular responses. UV radiation is connected to experimental conditions and organisms in which experiments were conducted, as well as the study type, while biomolecules such as genes, proteins, metabolites, and miRNAs are linked to their mapping to unique identifiers in standard databases.

Figure 2

Screenshots illustrate the user interface and navigation features of the UVREK website. (a) Screenshot of the UVREK homepage which has a structured layout for ease of navigation and access to compiled data. (b) BROWSE tab showing two primary ways of exploring the content: Browse by UV type or Browse by Biomolecules. (c) BROWSE tab enables users to explore UV radiation exposure induced biomolecule data on genes, proteins, metabolites, and miRNAs, with results conveniently displayed in tabular formats. Last column of the tabular output has links to the associated experimental information, and clicking the corresponding link will lead to the page shown in part (d). (d) Screenshot of the page containing detailed information on the experimental conditions and other metadata associated with the expression profile of the corresponding biomolecule. (e) Screenshot of the tabular column obtained as a result of clicking on the standardized identifiers of a biomolecule. The resulting page displays information on all studies within UVREK in which the biomolecule has been reported. (f) Screenshot of the SEARCH page enabling users to query the compiled biomolecules in UVREK database based on their identifiers, UV-specific details, PMIDs or experimental conditions. (g) Screenshot of the results obtained in a tabular format from an example query search.

Figure 3

Distribution of biomolecules induced by UV radiation exposure based on various criteria. (a) Distribution of biomolecules based on the tested organism namely, Homo sapiens (human), Mus musculus (mouse), or Rattus norvegicus (rat). (b) Distribution of biomolecules based on the type of UV radiation employed. (c) Distribution of biomolecules specific to humans based on the exposed organ or tissue types. (d) The top 10 genes based on the number of associated published studies or PubMed Identifiers (PMIDs) compiled in UVREK. The displayed distributions in this figure were obtained based on unique identifiers for genes, proteins, metabolites and miRNAs in UVREK as described in the Methods section.

Figure 4

Bar chart displaying the top 10 enriched gene ontology (GO) terms across three categories: biological process (depicted in blue), cellular component (depicted in purple), and molecular function (depicted in pink). Here, each horizontal bar gives the count of genes, among the 811 human genes associated with UV exposure, linked to the respective GO term, with statistical significance (p ≤ 0.01).

Figure 5

Pathway enrichment analysis of the 811 human genes induced by UV radiation exposure. (a) Bubble plot visualization of the 53 significantly enriched KEGG pathways in the input gene set, with bubble size reflecting significance (p-value) and placement of the bubble along the horizontal axis indicating gene overlap score (i.e., the level of overlap between input gene set and the pathway genes). Bubble colors correspond to KEGG BRITE classification at Level 2. (b) Pathway similarity network depicting relationships among the 53 enriched KEGG pathways in the input gene set. The similarity between any pair of pathways is quantified using the Jaccard index between the associated gene sets. Nodes are colored based on KEGG BRITE classification at Level 2, and pathways are connected if their Jaccard index exceeds 0.2.

Figure 6

Visualization of the protein–protein interaction (PPI) network corresponding to the human genes induced by UV exposure. Visualization of the largest connected component of the first-order PPI network built using (a) 811 human genes. (b) 648 skin-specific genes. Within the PPI networks, modules of size greater than 100 are colored differently from the rest of the nodes (colored in gray). In both PPI networks, the yellow, red and green colored modules correspond to the largest, second largest and third largest, respectively, in terms of number of constituent nodes, and these large modules capture highly interconnected genes with key functional interactions induced by UV exposure. Further, the labeled nodes indicate the top 10 hub genes in the PPI networks.

Figure 7

Systematic workflow for identifying published research articles containing information on the expression profiles of genes, proteins, metabolites, and miRNAs induced upon exposure to Ultraviolet (UV) radiation in humans and rodents.