Database
user guide.

Start here

Choose a database entry point

Use the homepage search for a gene or variant, or select a dataset from the navigation menu to browse its records. Tissue and cell-type filters are available only in tables with the corresponding input boxes; the homepage search is not a tissue or trait search.

A few controls you will use throughout

• Filter: on a dataset page, type into an input beneath a column heading and press Enter. Multiple column filters are applied together. Clear an input and press Enter to remove that filter. Text filters generally match the beginning of a value, rather than any occurrence within it; check the returned identifiers even when entering a full ID. Numeric inputs are not range expressions: do not enter conditions such as <0.05.
• Read a field: hover over the exclamation-mark icon beside a column name for its definition. A displayed column is not necessarily filterable or sortable.
• Sort: select a sortable heading to change the order. Only supported fields can be sorted; the request is applied on the server.
• Navigate: use the table pagination controls and horizontal scrollbar to see additional rows and columns.
• Export: “Export current page” saves the displayed page, not every matching record. Use Download for available bulk files.

Find your starting point

Home & search

Search from the homepage box or select the search icon at the end of the navigation menu on any page. Both open the same combined results page. The result groups shown depend on whether you search for a gene or a variant.

• Gene search (gene symbol or Ensembl gene ID): shows molecular expression, expression plots, JBrowse, QTL, fine mapping, independent signals, colocalization and TWAS.
• Variant search (for example, 1_38204722): shows QTL, fine mapping and independent signals. Molecular expression, expression plots, JBrowse, colocalization and TWAS are hidden for this search type.

A result group hidden for a variant search is expected interface behavior, not a query error. A visible group may still have no matching records in its dataset.

1. Enter a gene symbol such as HADH, an Ensembl gene ID such as ENSBTAG00000015728, or a variant ID such as 1_38204722 in the homepage box or navigation search panel. Use the shown capitalization and remove surrounding spaces.
2. Submit the search. A gene symbol is looked up to obtain its gene ID; a gene ID is used to query the result tables, with a gene-name lookup for the expression plot and JBrowse. Variant searches use the supplied identifier.
3. Expand a result group and select its dataset tab. Browse, sort supported columns or export the current page. The search identifier remains applied across the result tables; use the navigation search icon to start a new search. A match in one dataset does not imply a match in every other dataset.

Enlarge screenshot

Browse expression and fraction data

Molecular phenotypes

Gene expression

Find a gene and inspect its expression across tissues or tissue–cell-type combinations. The table includes genomic coordinates and separate chart controls for the two expression views.

1. Open Molecular phenotypes → Gene expression. Filter by Gene Id, Start, End or Chromosome, then press Enter. This table has no gene-symbol, tissue or cell-type filter; use the homepage search first if you only know a gene symbol.
2. Select the chart icon under Tissue Expression or Tissue_Celltype Expression in the required gene's row.
3. Both box-plot views display log2(TPM + 1) for non-zero TPM records and order groups by ascending median expression. They show distributions, not a single average expression value. Hover over the plot to inspect values; close the dialog to return to the table.

Enlarge screenshot

Enlarge screenshot

Cell type fraction

Open Molecular phenotypes → Cell type fraction. Filter by IID (sample), Tissue, Cell type or a numeric Fraction value, then press Enter. Select the icon in Plot to open a stacked bar chart for that row's tissue, with separate cell-type series. Zero-fraction records are excluded from both the table and plot queries.

Enlarge screenshot

Browse eQTL datasets

QTL / eQTL

Select the analysis context before filtering. The four eQTL entries are separate datasets, not interchangeable views of a single result.

The menu offers Bulk_eQTL, Cellstate_ieQTL, Celltype_ieQTL and Celltype_seQTL. Select one to open its table.

1. Choose a dataset under QTL → eQTL.
2. Enter a gene ID in pheno_id, a tissue in tissue_name, a variant ID in variant_id_x or a dbSNP ID in rsid, then press Enter. Try ENSBTAG00000015728 in Bulk_eQTL. The celltype_name column is displayed but has no filter input.
3. The table initially sorts pval_g1_x in ascending order. Select that heading to reverse the order, use the page controls to browse more matches, or export the current page. The other eQTL columns do not have sorting controls.

Enlarge screenshot

Browse fine-mapping records

Fine mapping

Browse the stored fine-mapping results for Bulk_eQTL or Celltype_seQTL. Selecting a dataset displays existing records; it does not run a new fine-mapping analysis.

1. Select QTL → Fine mapping and choose a dataset.
2. Filter using the SNP, rsid, gene or tissue_name input, then press Enter.
3. Results initially sort by PIP in descending order. Select the PIP heading to reverse it. cs, ct and annotation columns are available for reading, not filtering.

Enlarge screenshot

Browse independent-signal data

Independent signals

Browse the bulk or cell-type independent-signal tables. Each record includes gene, tissue and model fields.

1. Choose QTL → Independent, then Bulk_eQTL or Celltype_eQTL.
2. Filter by pheno_id, tissue, map_model or variant_id and press Enter.
3. Scroll to view beta_g1, beta_se_g1 and pval_g1. The af, beta_g1, beta_se_g1 and pval_g1 headings support sorting.

Enlarge screenshot

Browse model and result tables

TWAS

TWAS models

Open TWAS → TWAS models. Enter an Ensembl gene ID in gene or a gene symbol in gene_name, then press Enter. This table displays stored prediction-model metrics; it has no tissue or cell-type columns and does not train a new model.

Sortable fields are test_R2_avg, test_R2_sd, cv_R2_sd, rho_avg, rho_se, rho_avg_squared, zscore_pval, cv_rho_avg_squared, cv_zscore_pval and cv_pval_est. Hover over a field's exclamation-mark icon for its definition.

Enlarge screenshot

TWAS results

Open TWAS → TWAS results. Filter by gene or gene_name and press Enter. Scroll horizontally to read the columns displayed as tissue_celltype (underlying field tissue_ct) and trait_name; these columns have no filter inputs. You can sort pvalue, zscore, effect_size, pred_perf_r2, pred_perf_pval and best_gwas_p.

Enlarge screenshot

Browse colocalization records

Colocalization

The navigation provides separate gene-level and QTL-level result tables. Choose the corresponding entry before filtering records.

Gene-level results

Under Colocalization, choose Bulk_eGene, Cellstate_ieGene, Celltype_ieGene or Celltype_seGene. Filter by gene_id, tissue or celltype and press Enter. The traits column is display-only. Only the displayed res.summary heading (underlying field res_summary) supports sorting; hover over its information icon for the field definition.

Enlarge screenshot

QTL-level results

Choose Bulk_eQTL, Celltype_eQTL or Celltype_ieQTL to inspect the corresponding QTL records. Filter by pheno_id or variant_id and press Enter; tissue_name and trait_name are display-only. Scroll horizontally to view additional fields. The af, beta_g1, beta_se_g1 and pval_g1 headings support sorting.

Enlarge screenshot

Browse selection-signal data

Selection Signal

The current entry opens the fst10_cseQTL dataset, with identifier and annotation columns.

1. Filter by pheno_id, tissue_name, ct_name or the column labelled varaint_id, then press Enter.
2. You can also filter by ATAC_status, Chromatin_state, Enrichment or SNP_annotation.
3. Hover over an information icon for the column definition. No headings in this table provide sorting controls.

Enlarge screenshot

Work with the source files

Download

The Download page groups dataset file links. Some files are listed now; the page notes that the complete dataset will be made available upon publication.

1. Locate the relevant category and read the dataset description.
2. Expand a grouped category where an expansion control is provided.
3. Select the download icon for the required file. Links may open an external file host; large transfers depend on that host and your connection.

Enlarge screenshot

When you need a hand

FAQ & support

Screenshots are cropped from real interface captures at a 2560 × 1440 viewport. Enlarge an image to view it fitted to the window. They are examples, not a guarantee of current row counts or data availability.