Experiment / E5D9A3UOPStandard STARR-seq

HepG2 adapted STARR-seq allele activity screen

High-throughput functional dissection of noncoding SNPs with biased allelic enhancer activity for insulin resistance-relevant phenotypes

An episomal hSTARR-seq_ORI library containing both alleles of selected noncoding IR-associated SNPs was transfected into human Hep-G2 cells for 24 h. Three biological replicates were sequenced as plasmid input and poly(A)+ RNA output libraries.

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

Targeted variant library in the episomal hSTARR-seq_ORI self-transcribing reporter (Addgene plasmid 99296). Each tested SNP allele was represented by a 150-bp oligonucleotide containing a 120-bp genomic sequence centered on the SNP plus common adapters; 13-bp UMIs were added during input DNA processing or output RNA reverse transcription. The packaged table joins GEO UMI counts to the paper's voom per-allele activity calls (Table S5) and mpralm allele-contrast results (Table S6).

Processed data

50 rows per page. Click a cell to inspect its full value.

Visible columns (32 of 32)
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 · 32 definitions
snp_id
dbSNP identifier for the tested variant
allele_1
First allele in the paper's Table S6 A1 ordering; not relabeled as reference or alternate
allele_2
Second allele in the paper's Table S6 A2 ordering; not relabeled as reference or alternate
region
Cytogenetic region reported in Table S6
allele_1_input_rep1
Raw UMI count for allele 1 in input replicate 1
allele_1_input_rep2
Raw UMI count for allele 1 in input replicate 2
allele_1_input_rep3
Raw UMI count for allele 1 in input replicate 3
allele_1_output_rep1
Raw UMI count for allele 1 in output RNA replicate 1
allele_1_output_rep2
Raw UMI count for allele 1 in output RNA replicate 2
allele_1_output_rep3
Raw UMI count for allele 1 in output RNA replicate 3
allele_1_input_mean
Mean raw UMI count for allele 1 across the three input replicates
allele_1_output_mean
Mean raw UMI count for allele 1 across the three output replicates
allele_2_input_rep1
Raw UMI count for allele 2 in input replicate 1
allele_2_input_rep2
Raw UMI count for allele 2 in input replicate 2
allele_2_input_rep3
Raw UMI count for allele 2 in input replicate 3
allele_2_output_rep1
Raw UMI count for allele 2 in output RNA replicate 1
allele_2_output_rep2
Raw UMI count for allele 2 in output RNA replicate 2
allele_2_output_rep3
Raw UMI count for allele 2 in output RNA replicate 3
allele_2_input_mean
Mean raw UMI count for allele 2 across the three input replicates
allele_2_output_mean
Mean raw UMI count for allele 2 across the three output replicates
allele_1_activity_log2fc
Paper Table S5 log2 output/input activity effect for allele 1
allele_1_activity_fdr
Paper Table S5 FDR for the output-versus-input activity test for allele 1
allele_1_activity_type
Paper Table S5 per-allele call: eSNP, silencer, or inactive
allele_2_activity_log2fc
Paper Table S5 log2 output/input activity effect for allele 2
allele_2_activity_fdr
Paper Table S5 FDR for the output-versus-input activity test for allele 2
allele_2_activity_type
Paper Table S5 per-allele call: eSNP, silencer, or inactive
allelic_log2fc_a1_vs_a2
Paper Table S6 mpralm log2 allele effect, explicitly A1/A2; positive values favor allele 1
allelic_aveexpr
Paper Table S6 average expression statistic for the allele contrast
allelic_p_value
Paper Table S6 P value for the allele contrast
allelic_fdr
Paper Table S6 FDR-adjusted P value for the allele contrast
baaSNP
Paper Table S6 biased allelic enhancer activity call: YES or NO
min_count_all_samples
Minimum of the 12 raw UMI counts for both alleles across all input and output replicates; retained rows are at least 5

Quality control

The study demultiplexed paired reads by 8-bp replicate barcodes allowing up to two mismatches/deletions, filtered low-quality reads with fastp Q30 settings, extracted 13-bp UMIs and removed PCR duplicates, and retained only 100%-matched SNP sequences after Bowtie2 alignment. The paper excluded SNP alleles with fewer than 5 counts in any sample and reported replicate activity correlations of Pearson R² 0.91–0.97. For this package, a variant pair was retained only when both Table S6 alleles had at least 5 UMI counts in all 3 input and 3 output replicates (12 counts per pair); all 1,984 HepG2 pairs passed this package-level filter.

Curation notes

Samples are GSM5936694–GSM5936699 (input replicates 1–3 followed by output replicates 1–3). The paper treats HepG2 as a liver proxy; no treatment beyond reporter-library transfection was applied. The official Table S6 includes 393 baaSNP-positive and 1,591 baaSNP-negative contrasts in this packaged QC-passed set. GEO records the build as h19; this package normalizes that label to hg19. The paper separately uses hg38 conservation scores for a downstream deltaSVM analysis.

Cite OpenMPRA

Cite the OpenMPRA database. Include your access date because the collection changes over time.

Please also cite the source studies when using their data.