MCF-7 targeted ESR1 STARR-seq: vehicle versus E2
Comprehensive functional annotation of ESR1-driven enhancers in breast cancer reveals hierarchical activity independent of genomic and epigenomic contextsA custom oligonucleotide-capture library of ESR1-bound genomic regions from breast tumors and T-47D cells was tested in MCF-7 cells using a plasmid-based self-transcribing reporter. The processed table contains the authors' post-filtered 7,576 MCF-7 ESR1 binding sites and their vehicle/E2 enhancer activity statistics.
Processed tables are specific to each experiment. Column names, units, measurements, and table structure are not standardized across the database. Check this experiment’s column definitions and quality-control notes before comparing or combining data.
Perturbation & assay details
DMSO vehicle versus 10 nM 17β-estradiol (E2) for 6 h after 48 h hormone deprivation
A custom oligonucleotide capture library of 14,299 ESR1-bound genomic regions was cloned into the hSTARR-ORI plasmid (Addgene 99296), with captured inserts downstream of a minimal promoter. The library was transfected into hormone-deprived MCF-7 cells; after 24 h, cells received DMSO vehicle or 10 nM E2 for 6 h, and poly(A) STARR transcripts were sequenced as 150-bp paired-end Illumina NovaSeq 6000 reads. The processed scores are the authors' DESeq2 analysis of RNA-over-DNA signal and E2-versus-vehicle differential activity.
Processed data
50 rows per page. Click a cell to inspect its full value.
Visible columns (19 of 19)
| Row | |||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | |||||||||||||||||||
| 2 | |||||||||||||||||||
| 3 | |||||||||||||||||||
| 4 | |||||||||||||||||||
| 5 | |||||||||||||||||||
| 6 | |||||||||||||||||||
| 7 | |||||||||||||||||||
| 8 | |||||||||||||||||||
| 9 | |||||||||||||||||||
| 10 | |||||||||||||||||||
| 11 | |||||||||||||||||||
| 12 | |||||||||||||||||||
| 13 | |||||||||||||||||||
| 14 | |||||||||||||||||||
| 15 | |||||||||||||||||||
| 16 | |||||||||||||||||||
| 17 | |||||||||||||||||||
| 18 | |||||||||||||||||||
| 19 | |||||||||||||||||||
| 20 | |||||||||||||||||||
| 21 | |||||||||||||||||||
| 22 | |||||||||||||||||||
| 23 | |||||||||||||||||||
| 24 | |||||||||||||||||||
| 25 | |||||||||||||||||||
| 26 | |||||||||||||||||||
| 27 | |||||||||||||||||||
| 28 | |||||||||||||||||||
| 29 | |||||||||||||||||||
| 30 | |||||||||||||||||||
| 31 | |||||||||||||||||||
| 32 | |||||||||||||||||||
| 33 | |||||||||||||||||||
| 34 | |||||||||||||||||||
| 35 | |||||||||||||||||||
| 36 | |||||||||||||||||||
| 37 | |||||||||||||||||||
| 38 | |||||||||||||||||||
| 39 | |||||||||||||||||||
| 40 | |||||||||||||||||||
| 41 | |||||||||||||||||||
| 42 | |||||||||||||||||||
| 43 | |||||||||||||||||||
| 44 | |||||||||||||||||||
| 45 | |||||||||||||||||||
| 46 | |||||||||||||||||||
| 47 | |||||||||||||||||||
| 48 | |||||||||||||||||||
| 49 | |||||||||||||||||||
| 50 |
Filters apply to this table only. The CSV download contains the complete processed table; filtered rows are available through the API.
Column dictionary · 19 definitions
- element_id
- Unique STARR-seq library element identifier (source STARR_ID).
- chr_hg19
- Chromosome for the tested region in the source hg19 coordinate set.
- start_hg19
- Start coordinate for the tested region in hg19, as reported by the source table.
- end_hg19
- End coordinate for the tested region in hg19, as reported by the source table.
- chr_hg38
- Chromosome for the corresponding tested region in the source hg38 coordinate set.
- start_hg38
- Start coordinate for the tested region in hg38, as reported by the source table.
- end_hg38
- End coordinate for the tested region in hg38, as reported by the source table.
- base_mean
- DESeq2 baseMean: mean normalized STARR-seq signal across the analyzed samples/conditions.
- log2_fc_e2_vs_vehicle
- DESeq2 log2 fold change in STARR-seq RNA-over-DNA signal for E2 versus DMSO vehicle.
- lfc_se
- Standard error of the DESeq2 log2 fold-change estimate.
- wald_stat
- DESeq2 Wald test statistic for the E2-versus-vehicle contrast.
- p_value
- Raw DESeq2 P-value for the E2-versus-vehicle contrast.
- padj
- Benjamini-Hochberg adjusted P-value for the E2-versus-vehicle contrast.
- active_vehicle
- Source boolean flag indicating enhancer activity in the vehicle condition.
- active_e2
- Source boolean flag indicating enhancer activity in the E2 condition.
- constitutive_active_vehicle
- Source boolean flag for the constitutive-activity classification in vehicle.
- constitutive_active_e2
- Source boolean flag for the constitutive-activity classification in E2.
- activity_class
- Authors' final STARR-seq activity class: induced, constitutive, or inactive.
- neg_log10_padj
- Negative base-10 logarithm of padj, corresponding to the source log10adjpval column.
Quality control
Authors mapped raw FASTQ reads to hg19/GRCh37 with BWA v0.7.17, retained alignments with MAPQ >30 using SAMtools v1.14, converted aligned fragments to BEDPE, and counted fragments at target regions with BEDTools v2.30. Differential analysis used DESeq2 v1.30.1. From the 14,299-region capture library, they retained only regions overlapping MCF-7 ESR1 ChIP-seq peaks in full-medium conditions (GSE32222), yielding 7,576 unique sites; representation was checked against E2-stimulated MCF-7 ESR1 ChIP-seq (GSE54855). Published activity classes used these thresholds: inactive if the linear RNA/DNA fold change was <1 in both vehicle and E2; constitutive if E2/vehicle was <2 with adjusted P <0.05; and induced if E2/vehicle was >2 with adjusted P <0.05. The package retains all 7,576 rows in the authors' post-filtered Supplemental Table 1, including biologically inactive sites; no additional rows were removed because of biological inactivity.
Curation notes
This experiment reanalyzes previously generated STARR-seq data (EGA EGAD50000000015) and represents the MCF-7 arm used by the paper, not the separate Ishikawa arm present in the broader EGA dataset. The original 14,299-region capture library was reduced to 7,576 sites by the authors' MCF-7 ESR1 ChIP-seq overlap filter; the processed table contains 7576 data rows with 7576 unique element IDs and class counts of 580 induced, 4176 constitutive, and 2820 inactive, matching the publication. The table has both hg19/GRCh37 coordinates used for read mapping and the corresponding hg38 coordinates supplied by the authors. It contains region-level activity statistics rather than allele-specific variant effects; STARR_ID is a library element identifier, not an rsID. Raw reads remain controlled at EGA and are not included.