Experiment / E9TY631UYEpisomal Plasmid MPRA

Episomal MPRA of cancer non-coding variants across three human cancer cell lines

Widespread perturbation of ETS factor binding sites in cancer

A barcoded episomal MPRA library of 200-bp genomic fragments carrying reference and alternate cancer non-coding variants, together with control classes, was electroporated into Jurkat, HT-29, and SK-MEL-28 cells. The processed table contains the authors' MPRA-active-region results for 2,596 variants across the three cell-line contexts, with paired RNA/plasmid activity, allelic skew, tested sequences, and available TFA-BT promoter/TF annotations.

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: Not reported / not applicable
Reference genome: hg19
Design focus: Variant-focused
Region of interest: Not reported / not applicable

Perturbation & assay details

Basal / Untreated

Agilent oligos were 230 bp, containing 200 bp of genomic sequence and 15 bp of adapter sequence at each end. Unique 20-bp barcodes were added by PCR and the library was assembled into Addgene plasmid #109035; after AsiSI digestion and Gibson assembly, the 200-bp insert was placed directly upstream of a minimal TATA promoter driving GFP and the barcode was in the GFP 3′ UTR. The same library was tested by electroporation in human Jurkat (ATCC TIB-152), HT-29 (ATCC HTB-38), and SK-MEL-28 (ATCC HTB-72) cells. GFP RNA barcode libraries were sequenced with Illumina single-end 31-bp reads plus an 8-bp index, and the plasmid input was sequenced for barcode/oligo representation.

Processed data

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

Search table valuesFilter columnOperatorValue

Visible columns (56 of 56)

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 · 56 definitions

variant_id: Variant identifier in chromosome:position:reference:alternate format from the MPRA results workbook.
chromosome: Chromosome of the tested variant on hg19.
position_hg19: 1-based hg19 coordinate of the tested variant.
reference_allele: Reference allele carried by the reference reporter oligo.
alternate_allele: Alternate allele carried by the alternate reporter oligo.
variant_type: Allele-length classification; all packaged rows are single-nucleotide variants.
ncv_class: Cleaned NCV/library class from the source MPRA results; the source typo in the unobserved class was normalized.
source_ncv_class: NCV/library class exactly as supplied in Supplementary Data 2.
cell_line: Human cancer cell line in which the MPRA result was measured.
cell_line_biosample_curie: Cellosaurus CURIE for the row-level cell line: Jurkat CVCL:0367, HT-29 CVCL:0320, or SK-MEL-28 CVCL:0526.
reference_sequence: Forward 200-bp genomic sequence of the reference MPRA oligo.
alternative_sequence: Forward 200-bp genomic sequence of the alternate MPRA oligo.
reference_sequence_length_bp: Length in base pairs of the reference oligo sequence.
alternative_sequence_length_bp: Length in base pairs of the alternate oligo sequence.
reference_oligo_id: Reference oligo identifier from the MPRA sequence library workbook.
alternative_oligo_id: Alternate oligo identifier from the MPRA sequence library workbook.
associated_promoter_genes: Semicolon-separated genes associated with the variant's TFA-BT promoter annotations, when available.
associated_promoter_ids: Semicolon-separated internal promoter IDs from the TFA-BT annotation workbook, when available.
predicted_tf_names: Semicolon-separated transcription factors with predicted allele-dependent binding in the TFA-BT annotation workbook, when available.
predicted_tf_count: Number of unique predicted TF names in the TFA-BT annotations.
max_reference_motif_score: Maximum TFM-pvalue reference-allele motif score across available TFA-BT annotations.
max_alternate_motif_score: Maximum TFM-pvalue alternate-allele motif score across available TFA-BT annotations.
max_abs_motif_score_difference: Maximum absolute reference-versus-alternate motif-score difference (FIS) across available TFA-BT annotations.
predicted_binding_loss: Whether any available TFA-BT annotation flags a reference-only/lost binding case.
predicted_binding_gain: Whether any available TFA-BT annotation flags an alternate-only/gained binding case.
reference_plasmid_mean: Mean normalized plasmid-input count for the reference allele (source A.Ctrl.Mean).
reference_rna_mean: Mean RNA count for the reference allele (source A.Exp.Mean).
reference_activity_log2fc_rna_vs_plasmid: Reference-allele log2 RNA/plasmid activity estimate (source A.log2FC).
reference_activity_log2fc_se: Standard error of the reference-allele log2 RNA/plasmid activity estimate.
reference_activity_pvalue: Reference-allele activity p-value derived as 10^(-A.logP).
reference_activity_significant_fdr_0_01: Whether the reference allele passes the reported 1% BH-FDR activity threshold.
reference_activity_bh_fdr: Reference-allele Benjamini-Hochberg FDR derived as 10^(-A.logPadj_BH).
reference_activity_bonferroni_pvalue: Reference-allele Bonferroni-adjusted p-value derived as 10^(-A.logPadj_BF).
reference_activity_neglog10_p: Source negative-log10 reference-allele activity p-value.
reference_activity_neglog10_bh: Source negative-log10 reference-allele BH-adjusted activity p-value/FDR.
reference_activity_neglog10_bonferroni: Source negative-log10 reference-allele Bonferroni-adjusted activity p-value.
alternative_plasmid_mean: Mean normalized plasmid-input count for the alternate allele (source B.Ctrl.Mean).
alternative_rna_mean: Mean RNA count for the alternate allele (source B.Exp.Mean).
alternative_activity_log2fc_rna_vs_plasmid: Alternate-allele log2 RNA/plasmid activity estimate (source B.log2FC).
alternative_activity_log2fc_se: Standard error of the alternate-allele log2 RNA/plasmid activity estimate.
alternative_activity_significant_fdr_0_01: Whether the alternate allele passes the reported 1% BH-FDR activity threshold.
alternative_activity_pvalue: Alternate-allele activity p-value derived as 10^(-B.logP).
alternative_activity_bh_fdr: Alternate-allele Benjamini-Hochberg FDR derived as 10^(-B.logPadj_BH).
alternative_activity_bonferroni_pvalue: Alternate-allele Bonferroni-adjusted p-value derived as 10^(-B.logPadj_BF).
alternative_activity_neglog10_p: Source negative-log10 alternate-allele activity p-value.
alternative_activity_neglog10_bh: Source negative-log10 alternate-allele BH-adjusted activity p-value/FDR.
alternative_activity_neglog10_bonferroni: Source negative-log10 alternate-allele Bonferroni-adjusted activity p-value.
allelic_skew_log2fc_alt_vs_ref: Log2 alternate-versus-reference activity skew (source LogSkew; B.log2FC minus A.log2FC).
allelic_skew_pvalue: Allelic-skew p-value derived as 10^(-Skew.logP).
allelic_skew_fdr: Allelic-skew FDR derived as 10^(-Skew.logFDR); blank when the source FDR is missing.
allelic_skew_neglog10_p: Source negative-log10 allelic-skew p-value.
allelic_skew_neglog10_fdr: Source negative-log10 allelic-skew FDR.
allelic_skew_direction: Direction of the alternate allele relative to reference based on LogSkew.
allelic_skew_significant_fdr_0_05: Whether the source allelic-skew FDR is below 0.05; blank when no source FDR was reported.
mpra_active_fdr_0_01: Whether at least one allele passes the 1% MPRA activity FDR threshold; all packaged rows are true.
source_data2_row: 1-based worksheet row number in the downloaded Supplementary Data 2 MPRAs sheet, including its header row.

Quality control

The authors evaluated barcode coverage, replicate correlation, and separation of positive and negative controls. Reported mean detected barcode coverage was 215 per oligo in plasmid, 182 in Jurkat, 106 in HT-29, and 71 in SK-MEL-28; 97.6%, 96.8%, 94.0%, and 87.9% of oligos, respectively, had at least 10 barcodes. Replicate Pearson correlations were 0.96–1 for Jurkat and SK-MEL-28 and 0.85–0.98 for HT-29. Barcode counts were summed per oligo, replicates were median-normalized, RNA was independently centered to the 506 negative controls at log2 RNA/plasmid = 0 for each cell type, and DESeq2 negative-binomial modeling identified active oligos at FDR < 0.01. The authors tested reference-versus-alternate allelic skew on log-transformed RNA/plasmid ratios with a paired t-test and called emVars at FDR < 0.05. For this package, the author-provided MPRA results were retained only when at least one allele had source A/B.logPadj_BH >= 2 (equivalent to BH FDR <= 0.01 because those fields are -log10 values); all 5,838 rows pass this activity filter. Rows without significant allelic skew remain because they are valid active measurements, not failed sequencing QC. Conventional p-value/FDR columns were derived from the source -log10 significance fields, while the original -log10 fields are also retained.

Curation notes

This is one aggregate MPRA experiment because the study used one barcoded plasmid library across three cell-line contexts; biosample_id is null and the component Cellosaurus CURIE is preserved in each table row. The paper methods state six electroporation replicates, whereas the public GEO series contains six Jurkat RNA samples, five HT-29 RNA samples, five SK-MEL-28 RNA samples, and five plasmid-input samples; the GEO overall-design text says five Jurkat replicates. The paper methods state a 48-hour post-electroporation harvest, while the GEO summary says 24 hours; both source statements are retained as an ambiguity. The processed table is based on the author-provided Supplementary Data 2 active-region results, not a reimplementation of DESeq2 from the barcode matrix; the unnormalized GEO barcode-by-oligo counts remain in raw_data for reanalysis. Supplementary Data 7 supplies the joined 200-bp forward reference/alternate sequences, and Supplementary Data 1 annotations are present for the TFA-BT subset when exact variant keys match (3,641 of 5,838 rows). Non-significant allelic-skew rows were intentionally retained as valid active measurements. Missing source significance fields remain blank rather than being imputed.

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