Table 1 An overview of different MPRA approaches

 Barcoded MPRA [23, 31, 33, 36]

DNA sequences of interest are each synthesized in conjunction with a unique barcode and cloned into a plasmid upstream of a promoter, reporter gene, the unique barcode, and a poly-A tail. Sequences with regulatory activity drive expression of transcripts that include the barcode, such that barcode abundance in RNA extracted from transfected cells reflects regulatory element strength.

 STARR-seq [27]

Sequences of interest are cloned into a plasmid downstream of a minimal promoter (or, more recently, simply the origin of replication) and reporter gene and upstream of a poly-A tail. Sequences with regulatory activity drive expression of transcripts that include the sequence itself, such that the abundance of the focal sequence in RNA extracted from transfected cells reflects regulatory element strength.

 Lenti-MPRA [37]

Lentivirus is used to integrate MPRA libraries into the genome, thereby circumventing concerns that episomal reporter assays carried out via transient transfection may not reflect gene regulatory processes that take place in a native chromatin context. The cell-type range of lentivirus transduction is also much broader than transient transfection, opening the door to experiments in hard-to-transfect cell types.

 AAV MPRA [35]

MPRA libraries are packaged into an adeno-associated virus (AAV) for transfection. AAV is a nonpathogenic virus commonly used for gene therapy studies and permits transfection into a wide range of tissues, including post-mitotic tissues and tissues that are hard to transfect with traditional chemical or electrical methods. Unlike DNA delivered by lentivirus, the AAV-delivered DNA remains almost exclusively episomal.

 Saturation mutagenesis-based MPRA [38]

To test the functional effects of thousands of mutations in a candidate regulatory element, error-prone PCR is used to introduce sequence variation and to incorporate random sequence tags. These constructs are then assayed via the MPRA design to pinpoint SNPs that affect regulatory activity.

 STAP-seq [39]

Rather than measuring the activity of many candidate regulatory elements in the presence of a given minimal promoter, STAP-seq measures the responsiveness of many candidate promoters in the presence of a given element. Promoter candidates are cloned downstream of a strong enhancer and upstream of an ORF and poly-A tail. If a candidate fragment is capable of initiating transcription, it will produce reporter transcripts that start with the promoter candidate sequence wherever the TSS was initiated.

 UMI-STARR-seq [40]

This protocol introduces unique molecular identifiers (UMI) prior to post-transfection amplification of cell-extracted mRNA. The UMIs allow the researcher to account for PCR duplicates in downstream analyses, and are recommended especially for low complexity input libraries.

 ChIP-STARR-seq [41]

Open chromatin regions are incorporated into a DNA library, which is then assayed via STARR-seq.

 Pop-STARR-seq [42]

Regions of interest are amplified from DNA derived from many unique individuals. These genetically diverse products are then pooled and used as the input for STARR-seq.

 ATAC-STARR-seq [43]

Open chromatin regions are incorporated into a DNA library via ATAC-seq [44], and these elements are then assayed via STARR-seq. This design allows the researcher to preferentially test the activity of putative regulatory elements found within open chromatin in a given cell type.

 BiT-STARR-seq [45]

Oligos covering each of the alleles for a set of SNPs of interest are synthesized and incorporated into STARR-seq experiments to test for allele-specific expression. UMIs are also added during cDNA synthesis to account for PCR duplicates.

 mSTARR-seq [46]

STARR-seq style plasmid pools are constructed using a CpG-free reporter vector that retains the same functionality. Enzyme treatment is then used to create methylated and unmethylated versions of the plasmid pool, which can be assayed to identify regulatory sequences as well as methylation-dependent regulatory sequences.

 CapSTARR-seq [47]

Putative regulatory elements are selected from genomic DNA using hybridization capture-based target enrichment. Captured regions are then assayed via STARR-seq, allowing the researcher to test a targeted set of fragments without relying on oligo synthesis.