Experiment / E3FZESOK1Episomal Plasmid MPRA

Allelic eQTL/GWAS variant MPRA in GM12878 cells

Multiple causal variants underlie genetic associations in humans

A variant-focused library of human genomic fragments centered on selected variants was synthesized as reference/alternate allele pairs with random barcode assignment. Three independent plasmid transfections of GM12878 lymphoblastoid cells were harvested after 24 hours for reporter cDNA and plasmid-DNA barcode sequencing, with expression and allelic effects estimated by negative-binomial DESeq2 contrasts.

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

No exogenous treatment (basal GM12878 culture); 24 h post-electroporation

The article describes 150-bp variant-centered sequences with matched surrounding sequence for reference/alternate comparisons; the author design table records 180-bp synthesized oligo records. The library was cloned into a reporter plasmid and introduced by Neon electroporation. Reporter RNA was isolated as GFP mRNA, reverse-transcribed, and sequenced alongside plasmid DNA; barcode counts were clustered with Bartender v1.1, assigned to designed oligos, and summed across barcodes. The source count matrix includes 1KGTest, 1KGControl, and Sabeti pools plus additional non-Base haplotype records; this experiment represents the common Base allele-pair analysis.

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 package identifier composed of source pool, variant ID, coordinates, alleles, Base flag, and source-record occurrence.
pool
Author library pool label: 1KGTest, 1KGControl, or Sabeti.
variant_id
Author-designed variant identifier (varN).
gene_id
Gene or genes associated with the tested variant in the source library; NA is retained for control variants.
chromosome
Chromosome label from the source oligo table.
position_hg19
1-based variant position on the hg19 assembly, as supplied by GEO.
strand
Source-strand orientation for the oligo design.
reference_allele
Reference allele sequence in the tested pair; may be longer than one base for indels.
alternate_allele
Alternate allele sequence or comma-delimited set of alternate sequences in the source table.
variant_class
Derived class: SNV when reference and all listed alternate alleles are one base; otherwise indel_or_MNV.
n_alternative_alleles
Number of comma-delimited alternate allele sequences in alternate_allele.
haplotype_flag
Source haplotype context; all retained rows are Base after QC.
oligo_copy_index
Occurrence number within a duplicated source metadata key, preserving separate compact count records.
oligo_copy_count
Number of source records sharing the same pool, variant, coordinate, allele, and haplotype metadata.
qc_mean_raw_count
Arithmetic mean of the 12 raw plasmid-DNA and cDNA counts used for the author QC filter; all retained values are >150.
dna_ref_rep1
Raw plasmid-DNA count for the reference allele, replicate 1.
rna_ref_rep1
Raw reporter cDNA count for the reference allele, replicate 1.
dna_ref_rep2
Raw plasmid-DNA count for the reference allele, replicate 2.
rna_ref_rep2
Raw reporter cDNA count for the reference allele, replicate 2.
dna_ref_rep3
Raw plasmid-DNA count for the reference allele, replicate 3.
rna_ref_rep3
Raw reporter cDNA count for the reference allele, replicate 3.
dna_alt_rep1
Raw plasmid-DNA count for the alternate allele, replicate 1.
rna_alt_rep1
Raw reporter cDNA count for the alternate allele, replicate 1.
dna_alt_rep2
Raw plasmid-DNA count for the alternate allele, replicate 2.
rna_alt_rep2
Raw reporter cDNA count for the alternate allele, replicate 2.
dna_alt_rep3
Raw plasmid-DNA count for the alternate allele, replicate 3.
rna_alt_rep3
Raw reporter cDNA count for the alternate allele, replicate 3.
base_mean_expression
Author DESeq2 baseMean for the allele-independent reporter expression contrast.
log2_fold_change_expression
Author DESeq2 log2 fold-change for reporter expression relative to plasmid DNA, independent of allele.
expression_lfc_se
Standard error of log2_fold_change_expression from the author model.
expression_statistic
Author DESeq2 test statistic for the expression contrast.
expression_p_value
Nominal P-value for the expression contrast.
expression_p_adj
Benjamini-Hochberg adjusted P-value for the expression contrast.
base_mean_allelic
Author DESeq2 baseMean for the allele-dependent reporter activity contrast.
log2_fold_change_allelic
Author DESeq2 log2 fold-change for alternate versus reference reporter RNA/DNA activity; positive values indicate higher alternate activity.
allelic_lfc_se
Standard error of log2_fold_change_allelic from the author model.
allelic_statistic
Author DESeq2 test statistic for the allelic contrast.
allelic_p_value
Nominal P-value for the allelic contrast.
allelic_p_adj
Benjamini-Hochberg adjusted P-value for the allelic contrast.
expression_hit_fdr05
True when expression_p_adj is less than 0.05.
allelic_hit_fdr05
True when allelic_p_adj is less than 0.05.
joint_expression_allelic_hit_fdr05
True when both expression_p_adj and allelic_p_adj are less than 0.05.
allelic_effect_direction
Derived direction of the allelic log2 fold-change: higher_in_alternate, lower_in_alternate, or no_directional_effect.

Quality control

The author pipeline clustered barcodes with Bartender v1.1, assigned barcodes to parent oligos when perfectly matching, and summed barcode counts per oligo. The author model notebook retained only rows with haploFlag/haplo equal to Base and a mean greater than 150 across all 12 plasmid-DNA and cDNA count columns. From 46,699 raw count rows, 33,569 were Base rows and 30,522 passed the mean-count filter; all 13,130 non-Base haplotype rows and 3,047 low-count Base rows were excluded. The processed table contains all 30,522 QC-passing rows, including nonsignificant rows, and every retained row matched the author-supplied SupplementaryTableS4 statistics. The table separately flags 8,502 expression FDR < 0.05 rows, 2,734 allelic FDR < 0.05 rows, and 1,264 rows significant in both contrasts; the latter reproduces the paper's reported 1,264 allelic effects.

Curation notes

The processed table is one row per QC-passing Base oligo-pair record and retains source nonsignificant results for downstream fine-mapping. The compact count/results tables contain 85 duplicated metadata groups (88 extra rows), mostly arising from multiallelic designs; oligo_copy_index and oligo_copy_count preserve these records, while SupplementaryTableS2 contains the sequence-level allele labels needed to distinguish them. The source alternate_allele field is retained verbatim and may list multiple alternatives. The article reports 1,264 allelic effects in the joint expression-plus-allelic hit set; the supplied results contain 2,734 allelic FDR < 0.05 rows when considered independently. Primary coordinates are hg19 per GEO, whereas the retained mpraSites-hg38.txt is a separate downstream converted site list. Raw sequencing reads were not included.

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.