Drosophila genome-wide UMI-STARR-seq enhancer activity maps
DeepSTARR predicts enhancer activity from DNA sequence and enables the de novo design of synthetic enhancersA sheared genomic DNA library from Drosophila melanogaster strain y; cn bw sp was tested in S2 cells with DSCP developmental and Rps12 housekeeping core promoters in two biological replicates. The processed table contains merged developmental and housekeeping peak calls and their RNA/input enrichment measurements.
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
Basal / Untreated
Episomal Drosophila STARR-seq vectors carried either the developmental DSCP or housekeeping Rps12 core promoter. Genomic inserts averaged approximately 200 bp; reporter RNA received a 10-bp UMI, and RNA and DNA input were sequenced as paired-end UMI-STARR-seq libraries. Peak activity is reported as non-corrected STARR-seq/input enrichment, while the corrected lower 95% confidence-bound enrichment was used for enhancer calling.
Processed data
50 rows per page. Click a cell to inspect its full value.
Visible columns (13 of 13)
| 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 · 13 definitions
- element_id
- Unique screen-specific identifier for a called genomic enhancer peak.
- locus_id
- Genomic locus represented as chromosome:summit coordinate; the same locus can occur in both promoter screens.
- chrom
- Drosophila dm3 chromosome or chromosome arm containing the peak summit.
- summit
- Peak summit coordinate from the merged GEO peak call.
- screen
- Core-promoter screen: developmental (DSCP) or housekeeping (Rps12).
- starrseq_reads_summit
- Merged STARR-seq reporter-read count at the peak summit.
- input_reads_summit
- Merged plasmid DNA input-read count at the peak summit.
- starrseq_reads_total
- Total merged STARR-seq reporter reads in the corresponding screen.
- input_reads_total
- Total merged plasmid DNA input reads in the corresponding screen.
- enrichment_starrseq_over_input
- Non-corrected STARR-seq/input enrichment at the peak.
- corrected_enrichment_lower_95ci
- Conservative lower bound of the 95% confidence interval for enrichment, used for peak QC.
- p_value
- Hypergeometric peak-call p-value.
- qc_pass
- TRUE for peaks retained after the author-reported p-value and corrected-enrichment filters.
Quality control
Author-reported processing mapped paired-end reads to dm3 with Bowtie, retained mappings with at most three mismatches and insert size at most 2 kb, collapsed RNA reads at identical positions by 10-bp UMI allowing one mismatch, retained 150-250-bp fragments, excluded chromosomes U, Uextra and mitochondrial DNA, and pooled the two biological replicates per promoter. Enhancer peaks were called with hypergeometric p-value <= 0.001 and corrected enrichment over input >3; the GEO peak files and this table retain only entries meeting those thresholds.
Curation notes
The paper calls this a genome-wide UMI-STARR-seq assay; the processed table uses the two GEO merged peak files (11,658 developmental and 7,062 housekeeping peaks), one row per screen-specific peak. The generic S2 cell line description was mapped to canonical Schneider 2 in Cellosaurus (CVCL:Z232); the paper does not specify a more precise S2 isolate. The complete author sequence-level activity/prediction table is retained in raw_data for users who need 249-bp sequences and model predictions.