​

Genomic Elements

The IGVF Catalog contains genomic elements, which are potential functional regions for regulating gene expression. These elements are identified/tested by functional assays (e.g., MPRA), association studies (e.g., caQTLs), and computational predictions (e.g., the ENCODE-rE2G model). Data sources include:

• ENCODE Candidate Cis-Regulatory Elements (cCREs): A unified set of genomic elements predicted to have regulatory potential based on a combination of experimental data (e.g., ChIP-seq, DNase-seq) from the ENCODE project.

Additional genomic elements are added when studied in specific contexts, for example accessible elements used in predictive modeels or elements tested in an MPRA experiment or CRISPR screen

​

Quickly reach the right data type

• Region pages now start with a set of cards for Coding, Noncoding, Molecular Networks, Phenotypes, and Allele Frequencies so you can jump straight to the matching section of the page.
• Select the card that matches the question you have—allele frequency summaries, regulatory evidence, network context, or phenotypic links—and the page scrolls there for you.
• The cards stay easy to tap on phones and tablets, making it simple to switch between region-level tables during review sessions.

​

Stay oriented and read sources clearly

• A slim coverage bar under the page title mirrors the home view so you can immediately see how much coding, regulatory, network, phenotype, and frequency information exists for the region you opened.
• Section headings now call out whether the rows come from datasets (such as FAVOR or MaveDB experimental releases) or predictions (like ENCODE E2G). The label sits beside the source name so you know when you are looking at measured activity versus model output.

​

Region overview at a glance

• A summary card now opens each region page, combining the region length, whether it overlaps coding sequence or sits in an intergenic area, and the nearest gene in one sentence.
• When a region does not cover coding sequence, the summary states this plainly so you know to focus on noncoding evidence instead.

​

Coding Variants section appears only when relevant

• The Coding Variants section now shows only when the region overlaps coding sequence; otherwise the page displays the message “This region does not overlap any coding sequence.”
• This keeps noncoding regions focused on regulatory and biosample evidence without empty tables.

​

“Measured activity” table (renamed and simplified)

• The biosample evidence table is now titled Measured activity and keeps just the columns researchers requested: Biosample, Term, Description, Activity, and Source.
• Each row highlights the biosample context, a short plain-language description, the activity readout, and where the measurement comes from.

​

Coming soon notice

• A banner on the region page now lets you know that CRISPR and MPRA experiments and noncoding variant predictions are on the way, so you can anticipate upcoming evidence types.

​

Clear guidance for very large regions

• If you enter a region that exceeds the supported size, the page now shows an explanatory alert with the size limit and suggestions to narrow the range, instead of failing silently.

​

Genomic Elements-Genes Edges

​

Genomic Elements-Biosamples Edges

​

Variants-Genomic Elements Edges

(also on variants page)

• External caQTL studies: variants associated with genomic elements function from external studies (PMID:34038741, PMID:34017130) and AFGR(African Functional Genomics Resource)
• IGVF BlueSTARR: computational model predicts variants affecting the regulatory function in ENCODE cCREs (Candidate Cis-Regulatory Elements)
• IGVF MPRA

​

Enhancer-Gene Prediction(s)

This table reports whether the query region contains predicted enhancers from the ENCODE-rE2G model, and their target genes and cell types. Each row reports one predicted enhancer, target gene, and cell type. Currently, this table includes predictions from the ENCODE-rE2G model across 1700+ ENCODE biosamples (see Gschwind et al. bioRxiv 2023) The table is initially sorted by Score in descending order, showing the strongest predictions first.