LNCaP episomal MPRA of prostate-cancer somatic SNVs in transcription-factor tumor cistromes
Role of Noncoding Single Nucleotide Variants on Transcription Factors’ activity in CancerA variant-focused barcoded reporter library tested 710 somatic prostate-cancer SNVs in regulatory elements bound by FOXA1, AR, and/or HOXB13 tumor cistromes. Reference and mutant 87-bp elements were assayed in untreated LNCaP cells with an episomal luc2 reporter, and the processed package summarizes the 13 unique functional SNVs displayed in the study’s MPRA result figure.
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
The library used 146-mer oligonucleotide constructs containing an 87-bp variable sequence with the reference or somatic allele centered between 43-bp natural flanks, universal primer sites, KpnI/XbaI sites, and a 10-bp barcode. Oligos were cloned into pMPRA1 (Addgene #49349), and a minimal luc2 promoter from pMPRAdonor2 (Addgene #49353) was added to make the luc2+ library. LNCaP cells were transfected with 20 micrograms of library across ten electroporations, pooled, recovered for 24 hours, and harvested for genomic DNA and mRNA/cDNA tag sequencing on HiSeq 2500. MPRA was performed in triplicate. The dissertation methods specify 15 barcodes per allele; the final Cancer Cell article reports a different barcode count, so the discrepancy is preserved in agent_comments.
Processed data
50 rows per page. Click a cell to inspect its full value.
Visible columns (15 of 15)
| Row | |||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | |||||||||||||||
| 2 | |||||||||||||||
| 3 | |||||||||||||||
| 4 | |||||||||||||||
| 5 | |||||||||||||||
| 6 | |||||||||||||||
| 7 | |||||||||||||||
| 8 | |||||||||||||||
| 9 | |||||||||||||||
| 10 | |||||||||||||||
| 11 | |||||||||||||||
| 12 | |||||||||||||||
| 13 |
Filters apply to this table only. The CSV download contains the complete processed table; filtered rows are available through the API.
Column dictionary · 15 definitions
- figure_variant_label
- Variant label exactly as printed beside the Figure 3.7c boxplot.
- variant_id
- Normalized variant identifier in chromosome:position:reference>alternate form.
- chromosome
- Chromosome from the figure label, normalized with a chr prefix.
- position_hg19
- 1-based genomic position from the figure label, interpreted on hg19 as used by the study.
- reference_allele
- Reference allele shown before the hyphen in the figure label.
- alternate_allele
- Somatic variant allele shown after the hyphen in the figure label.
- tf_cistrome_membership
- Tumor-cistrome membership decoded from the figure’s blue FOXA1, red AR, and green HOXB13 color bar.
- activity_direction
- Qualitative direction of the mutant/variant construct relative to the reference construct as shown by the Figure 3.7c boxplot distributions.
- mpra_significant_q_lt_0_05
- Boolean indicating the study-reported Benjamini–Hochberg q < 0.05 functional call.
- reported_q_value
- Exact adjusted q-value; blank because the accessible source reports the threshold and plotted distribution but not the per-SNV values.
- reported_activity_effect
- Exact numeric MPRA activity effect; blank because only boxplots were publicly available for these rows.
- biological_replicates
- Number of MPRA biological replicates reported in the methods.
- tested_screen_count
- Total number of somatic SNVs in the MPRA screen after the study’s reported construct/QC handling.
- source_figure
- Figure and panel supplying the row-level result.
- quality_control_status
- Package QC status; rows are restricted to the study-reported functional set shown in the figure.
Quality control
The study normalized DNA and RNA barcode counts per million, added a pseudocount of 1, log2 transformed the values, removed barcodes with fewer than 1 transformed DNA count, assessed replicate Pearson correlations, quantile-normalized and combined replicate activity values, compared reference and mutant constructs with a Mann–Whitney U test, and omitted variants with fewer than 10 reference or mutant constructs. The reported screen retained 710 tested SNVs and called functional effects at Benjamini–Hochberg q < 0.05. The processed table is intentionally limited to the 13 unique functional SNVs displayed in Figure 3.7c; all rows are marked as passing the study’s reported QC/significance criteria. Exact barcode counts, effect sizes, and q-values were not available in the accessible data and are left blank.
Curation notes
This is a genuine episomal plasmid MPRA, not a ChIP-seq or luciferase-only result. The associated public GEO/EGA accessions contain ChIP-seq data used to define cistromes, while no public barcode-level MPRA matrix or oligo sequence library was found. The table is consequently a transparent, figure-derived summary of the 13 unique significant MPRA SNVs rather than a reconstructed count matrix. The parent study is an open-access thesis; its Chapter 3 was published as Mazrooei et al., Cistrome Partitioning Reveals Convergence of Somatic Mutations and Risk Variants on Master Transcription Regulators in Primary Prostate Tumors, Cancer Cell 2019, DOI 10.1016/j.ccell.2019.10.005.