Experiment / E9RIH2ULRWhole-Genome STARR-seq (WHG-STARR-seq)

Whole-genome STARR-seq of MDA-MB-231 cis-regulatory elements under circadian/proteasome drug treatments

Targeting circadian transcriptional programs through a cis-regulatory mechanism in triple negative breast cancer

The authors sonicated MDA-MB-231 genomic DNA to approximately 700-bp fragments, cloned the library into the hSTARR-ORI episomal reporter vector, and transfected it into MDA-MB-231 cells. Four treatment groups (DMSO, SHP1705, MG132, and SHP1705 plus MG132) were sequenced as poly(A)+ reporter RNA and analyzed by MACS2 peak calling; GEO labels the samples as 8-hour treatments, while the STARR-seq method states that RNA was extracted 6 hours after transformation.

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 control; SHP1705 10 µM; MG132 100 nM; SHP1705 10 µM plus MG132 100 nM; GEO sample annotations designate 8-hour treatment, but the paper's STARR-seq method states RNA harvest 6 hours after transformation

A fragmented genomic DNA library from MDA-MB-231 cells (approximately 700 bp average) was inserted into the hSTARR-ORI vector (Addgene 99296), expanded in bacteria, and transfected episomally with Lipofectamine 3000. Poly(A)+ reporter RNA was collected 6 hours after transformation, reverse-transcribed with a library-specific primer, and sequenced on an Illumina HiSeq 4000. Reads were adapter-trimmed, paired-read and FastQC filtered, aligned to hg38 with Bowtie2, and called as peaks with MACS2. The study reports more than 10 million unique reads per group, but does not provide accurate input fragment copy numbers or barcode-level DNA/RNA counts; therefore, condition comparisons in the generated table are normalized peak-pileup descriptors rather than quantitative reporter fold changes.

Processed data

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

Visible columns (43 of 43)
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 · 43 definitions
element_id
Stable identifier for a consensus locus formed by the overlap union of source MACS2 peaks.
chrom
Chromosome or hg38 contig for the consensus locus.
start
Inclusive MACS2 coordinate of the consensus locus start, as reported by the source files.
end
Inclusive MACS2 coordinate of the consensus locus end, as reported by the source files.
length
Inclusive consensus locus width, calculated as end-start+1.
conditions_present
Treatment groups with at least one overlapping source peak, in DMSO|SHP1705|MG132|COMBO order.
condition_present_count
Number of treatment groups with an overlapping source peak.
dmso_peak_name
Source DMSO MACS2 peak name(s) overlapping the consensus locus.
dmso_peak_count
Number of overlapping DMSO MACS2 peak calls.
dmso_summit
Absolute summit coordinate of the overlapping DMSO call with the greatest pileup.
dmso_pileup
Sum of DMSO MACS2 pileup values across overlapping source calls.
dmso_normalized_pileup_per_100k
DMSO pileup normalized to the total retained DMSO peak pileup and scaled to 100000; not transcript TPM.
dmso_fold_enrichment
Maximum MACS2 fold enrichment among overlapping DMSO calls.
dmso_neg_log10_pvalue
Maximum MACS2 -log10(p-value) among overlapping DMSO calls.
dmso_neg_log10_qvalue
Maximum MACS2 -log10(q-value) among overlapping DMSO calls.
shp1705_peak_name
Source SHP1705 MACS2 peak name(s) overlapping the consensus locus.
shp1705_peak_count
Number of overlapping SHP1705 MACS2 peak calls.
shp1705_summit
Absolute summit coordinate of the overlapping SHP1705 call with the greatest pileup.
shp1705_pileup
Sum of SHP1705 MACS2 pileup values across overlapping source calls.
shp1705_normalized_pileup_per_100k
SHP1705 pileup normalized to the total retained SHP1705 peak pileup and scaled to 100000; not transcript TPM.
shp1705_fold_enrichment
Maximum MACS2 fold enrichment among overlapping SHP1705 calls.
shp1705_neg_log10_pvalue
Maximum MACS2 -log10(p-value) among overlapping SHP1705 calls.
shp1705_neg_log10_qvalue
Maximum MACS2 -log10(q-value) among overlapping SHP1705 calls.
mg132_peak_name
Source MG132 MACS2 peak name(s) overlapping the consensus locus.
mg132_peak_count
Number of overlapping MG132 MACS2 peak calls.
mg132_summit
Absolute summit coordinate of the overlapping MG132 call with the greatest pileup.
mg132_pileup
Sum of MG132 MACS2 pileup values across overlapping source calls.
mg132_normalized_pileup_per_100k
MG132 pileup normalized to the total retained MG132 peak pileup and scaled to 100000; not transcript TPM.
mg132_fold_enrichment
Maximum MACS2 fold enrichment among overlapping MG132 calls.
mg132_neg_log10_pvalue
Maximum MACS2 -log10(p-value) among overlapping MG132 calls.
mg132_neg_log10_qvalue
Maximum MACS2 -log10(q-value) among overlapping MG132 calls.
combo_peak_name
Source SHP1705-plus-MG132 combination MACS2 peak name(s) overlapping the consensus locus.
combo_peak_count
Number of overlapping combination MACS2 peak calls.
combo_summit
Absolute summit coordinate of the overlapping combination call with the greatest pileup.
combo_pileup
Sum of combination MACS2 pileup values across overlapping source calls.
combo_normalized_pileup_per_100k
Combination pileup normalized to the total retained combination peak pileup and scaled to 100000; not transcript TPM.
combo_fold_enrichment
Maximum MACS2 fold enrichment among overlapping combination calls.
combo_neg_log10_pvalue
Maximum MACS2 -log10(p-value) among overlapping combination calls.
combo_neg_log10_qvalue
Maximum MACS2 -log10(q-value) among overlapping combination calls.
shp1705_vs_dmso_log2_normalized_pileup_ratio
Descriptive log2 ratio of SHP1705 versus DMSO normalized peak pileup, reported only when both conditions have a peak at the locus.
mg132_vs_dmso_log2_normalized_pileup_ratio
Descriptive log2 ratio of MG132 versus DMSO normalized peak pileup, reported only when both conditions have a peak at the locus.
combo_vs_dmso_log2_normalized_pileup_ratio
Descriptive log2 ratio of combination versus DMSO normalized peak pileup, reported only when both conditions have a peak at the locus; not a calibrated reporter effect.
combo_vs_dmso_presence_pattern
Whether the consensus locus has a peak call in both DMSO and COMBO, only COMBO, or only DMSO.

Quality control

The authors trimmed adapters with Trimmomatic, retained paired reads, checked sequencing quality with FastQC, aligned QC-passed reads to hg38 with Bowtie2, and generated the deposited peak files with MACS2 at q=0.01. For packaging, comment/header lines were removed and each source row was required to have a valid hg38-style contig, positive inclusive interval length matching end-start+1, positive pileup, and -log10(q-value) >= 2. All 270743 source peak rows passed these checks; no peak rows were removed. The STARR-seq series has one sample per treatment, so replicate concordance and replicate-level significance cannot be assessed.

Curation notes

The publication calls this a massive parallel reporter assay, while the detailed method is an episomal whole-genome STARR-seq screen. The library is made from sheared MDA-MB-231 genomic DNA and is not a designed SNP/indel library; no rsIDs, alternative alleles, barcodes, matched input-DNA counts, or biological replicates are available for this STARR-seq arm. The GEO sample titles/designations say 8-hour treatment, whereas the paper's STARR-seq method says total RNA was extracted six hours after transformation; both annotations are preserved here. The paper explicitly states that input copy number could not be accurately defined and that precise cross-treatment fold-change analysis was therefore not reliable. The table consequently uses the exact deposited MACS2 peak evidence, merges overlapping calls across treatments into consensus loci, and labels normalized-pileup ratios as descriptive only. Valid hg38 alternate/random contigs were retained. The GEO peak files and the authors' GitHub peak files are byte-identical.

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.