Experiment / E7VS7DXHGIntegrated lentiMPRA

14,999-tile erythroid enhancer integrated lentiMPRA in HUDEP-2

Large-scale discovery of potent, compact and erythroid specific enhancers for gene therapy vectors

A 14,999-element library of overlapping tiles from 5,393 erythroid DNase-I hypersensitive sites was cloned upstream of a 169-bp minimal human β-globin promoter driving GFP in an integrating lentiviral vector. WT HUDEP-2 cells were transduced at MOI 0.4, sorted five days later into GFP-low, medium, and high bins, and the integrated tile DNA was sequenced to estimate enhancer activity.

Download processed table (CSV)Raw source data API documentation

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.

Organism: Human
Taxonomy ID: NCBITaxon:9606
Biosample: CVCL:VI06
Reference genome: GRCh38
Design focus: Region-focused
Region of interest: Not reported / not applicable

Perturbation & assay details

Untreated WT HUDEP-2 cells

Modified FACS-based lentiMPRA with a clinically oriented reporter architecture: each genomic tile was placed upstream of a minimal β-globin promoter and GFP ORF, and a C1 enhancer-blocker/barrier insulator flanked the integrated cassette. HUDEP-2 cells were transduced at MOI 0.4 and three equiproportional GFP-positive bins were collected (approximately 5% of live singlets per bin; mean MFIs about 750, 3,000, and 15,000). DNA insert counts were generated by HiSeq 4000 sequencing and the paper estimated tile effects with a MAUDE-adapted negative-binomial maximum-likelihood model.

Processed data

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

Search table valuesFilter columnOperatorValue

Visible columns (29 of 29)

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

Page 1 · 50 rows · More results available

Filters apply to this table only. The CSV download contains the complete processed table; filtered rows are available through the API.

Column dictionary · 29 definitions

element_id: Unique genomic tile identifier in chr:start-end format.
chrom: Chromosome of the tile on GRCh38.
start: Inclusive GRCh38 tile start coordinate.
end: Inclusive GRCh38 tile end coordinate.
element_length_bp: Tile length in base pairs, calculated as end - start + 1.
annotation: Genic feature annotation supplied by the authors.
nearest_gene: Nearest HGNC gene symbol supplied by the authors.
activity_score: Author-reported maximum-likelihood tile effect score for GFP expression; positive values indicate higher and negative values lower activity under the model.
activity_fit_distribution_quantile: Author-reported quantile of the fitted activity-score distribution used to identify extreme potential enhancers and silencers.
phastCons: Author-supplied phastCons conservation score.
motif_archetypes: Semicolon-delimited author-supplied transcription-factor motif archetypes overlapping the tile.
replicate_E_gfp_high_count: DNA-sequencing read count for replicate E, GFP-high bin; GEO NA was converted to 0.
replicate_E_gfp_medium_count: DNA-sequencing read count for replicate E, GFP-medium bin; GEO NA was converted to 0.
replicate_E_gfp_low_count: DNA-sequencing read count for replicate E, GFP-low bin; GEO NA was converted to 0.
replicate_E_all_gfp_count: DNA-sequencing read count for replicate E, all-GFP reference sample; GEO NA was converted to 0.
replicate_F_gfp_high_count: DNA-sequencing read count for replicate F, GFP-high bin; GEO NA was converted to 0.
replicate_F_gfp_medium_count: DNA-sequencing read count for replicate F, GFP-medium bin; GEO NA was converted to 0.
replicate_F_gfp_low_count: DNA-sequencing read count for replicate F, GFP-low bin; GEO NA was converted to 0.
replicate_F_all_gfp_count: DNA-sequencing read count for replicate F, all-GFP reference sample; GEO NA was converted to 0.
count_matrix_has_NA: True when at least one of the eight original GEO count cells for the tile was NA.
replicate_E_gfp_high_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-high E count over the E all-GFP reference, using a 0.5 pseudocount.
replicate_E_gfp_medium_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-medium E count over the E all-GFP reference, using a 0.5 pseudocount.
replicate_E_gfp_low_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-low E count over the E all-GFP reference, using a 0.5 pseudocount.
replicate_F_gfp_high_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-high F count over the F all-GFP reference, using a 0.5 pseudocount.
replicate_F_gfp_medium_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-medium F count over the F all-GFP reference, using a 0.5 pseudocount.
replicate_F_gfp_low_log2_enrichment_vs_all: Library-size-normalized log2 enrichment of the GFP-low F count over the F all-GFP reference, using a 0.5 pseudocount.
mean_gfp_high_log2_enrichment_vs_all: Mean of the E and F GFP-high log2 enrichments.
mean_gfp_medium_log2_enrichment_vs_all: Mean of the E and F GFP-medium log2 enrichments.
mean_gfp_low_log2_enrichment_vs_all: Mean of the E and F GFP-low log2 enrichments.

Quality control

The paper reports recovery of 14,668 of 14,999 designed tiles (97.8%) at approximately 800 unique integrations per element, with robust replicate concordance at that coverage. The deposited reads were trimmed, adapter-trimmed, aligned to GRCh38, filtered with samtools view -q 1 -f 64, converted to BED, and counted when overlapping a reference tile by at least 90%. This package retains only the 14,668 tile IDs present in the GEO count matrix and collapses 103 identical duplicate annotation rows in Supplementary Data 2 to one row per tile ID. No extra read-count threshold was imposed because none was reported for the final activity-score table. GEO count cells marked NA were encoded as zero for the derived enrichment columns and flagged in count_matrix_has_NA.

Curation notes

The eight count-matrix columns were mapped to GEO samples using the order and titles in GSE252159_family.soft.gz. The published activity_score and fitted-distribution quantile were preserved rather than recomputed. This experiment is a protein-readout modification of a lentiMPRA: the regulatory insert is upstream of the reporter promoter, and DNA tile abundance is measured after FACS binning. The full-DHS mini-library is packaged separately as experiment E0CMEOWJN.

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

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

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