Fig. 1

Overview of MPRA workflows. A (1) In the barcoded MPRA design, candidate regions of interest are synthesized via large-scale oligosynthesis. (2) The single-stranded DNA is paired with a unique barcode and converted to double-stranded DNA via PCR. (3) The barcoded DNA fragments are then cloned into an empty MPRA reporter vector. Next, the plasmid library is linearized between the barcode and the candidate query sequence, and (4) a minimal promoter (often SCP1) and open reading frame are inserted. (5) This plasmid pool is delivered via transfection (or infection if viral delivery is used) into the desired cell type, where (6) functional regulatory elements sequences will interact with the promoter to drive transcription of the ORF and the barcode, which is incorporated into each transcript’s 3′UTR. Finally, RNA is harvested from the cells, and (7) mRNA is sequenced to measure post-experiment barcode abundance, along with DNA fragments from the empty MPRA reporter vector step to identify query sequence-barcode associations. B (1) In the classic STARR-seq design, sequencing adapters as well as sequences complementary to the STARR-seq vector are added to DNA fragments of interest. (2) This fragment pool is then cloned into the STARR-seq vector upstream of a 3′ poly-adenylation signal and downstream of a promoter and synthetic intron (to differentiate spliced mSTARR-seq RNA transcripts from plasmid DNA in downstream PCRs). (3) After delivery into a cell line, (4) inserts that possess regulatory activity interact with the promoter to drive expression of the insert itself. Finally, RNA is harvested from the cells, and (5) mRNA is sequenced to measure post-experiment fragment abundance, along with DNA fragments from the pre-transfection (or pre-infection) plasmid pool to control for variation in input. See Fig. 2 for further information on data analysis