This project is described in the preprint:

www.biorxiv.org/content/10.1101/2021.03.16.435663v1

 

As of today, four high-quality systematic yeast PPI datasets are available via this web portal. 

 

Systematic efforts

Uetz-screen: The first mapping effort of the yeast protein interactome (Uetz et al. Nature 2000). Only the subset of PPIs detected in the high-throughput screen of an activation library (692 PPIs), are reported on this web-portal. 

 

Ito-core: The second mapping effort of the yeast interactome (Ito et al. PNAS 2001). Only PPIs from the core data (more than 3 IST hits, 841 PPIs) are reported on this web-portal. 

 

CCSB-YI1: The third systematic map (Yu et al. Science 2008), constructed at CCSB, identified 1,809 binary PPIs after three Y2H screens.

 

YeRI: Our latest mapping effort led to the identification of 1,910 PPIs after testing ~99% of the yeast proteome. This new map was constructed using a novel Y2H assay version, which uses high-copy number 2µ vectors instead of centromeric vectors used by the previously published maps (Uetz library, Ito core and YI-I).

 

The union of these four systematically generated PPI datasets is refered to as ABBI-21, for atlas of binary biophysical interactions, generated in 2021.

 

Y2H screen description of YeRI

Y2H bait and prey libraries: Yeast ORFs from the Harvard Institute of Proteomics collection were transferred by Gateway recombinational cloning (Invitrogen) into Y2H destination vectors pDEST-DB-QZ212 to generate Gal4 DNA binding domain hybrid proteins (DB-ORF) and pDEST-AD-QZ213 to generate Gal4 activation domain hybrid proteins (AD-ORF) as described previously (Dreze et al, Methods in Enzymology 2010).

 

Yeast strains: Yeast strains Y8930 (Genotype: MATa leu2-3,112 trp1-901 his3Δ200 ura3-52 gal4Δ gal80Δ GAL2::ADE2 GAL1::HIS3@LYS2 GAL7::lacZ@met2 cyh2R) and Y8800 (Genotype: MATa leu2-3,112 trp1-901 his3Δ200 ura3-52 gal4Δ gal80Δ GAL2::ADE2 GAL1::HIS3@LYS2 GAL7::lacZ@met2 cyh2R) were transformed with individual DB-ORF and AD-ORF constructs respectively.

 

Vector details:

 

Y2H screens and assay versions: Fresh overnight cultures of individual Y8930:DB-ORF yeast strains were mated against Y8800:AD-ORF mini-libraries containing ~700 different Y8800:AD-ORF yeast strains. After overnight growth at 30ºC in liquid rich medium (YEPD), mated yeast cells were transferred into SC-Leu-Trp liquid media to select for diploids. After overnight incubation at 30ºC diploid yeast cells were spotted onto SC-Leu-Trp-His+1mM 3AT solid media to select for activation of the GAL1::HIS3 reporter gene.

 

Acknowledgments

CCSB interactome mapping and ORFeome cloning efforts are supported by federal grants from the National Human Genome Research Institute.

Y2H bait and prey libraries: Yeast ORFs from the Harvard Institute of Proteomics collection were transferred by en masse Gateway recombinational cloning (Invitrogen) into high complexity barcoded Y2H destination vectors to generate Gal4 DNA binding domain hybrid proteins (DB-ORF) and Gal4 activation domain hybrid proteins (AD-ORF) as described previously (Yachie et al, MSB, 2016). High-quality ORF-barcode pairs were identified using the previously described KiloSeq procedure (Weile et al, MSB, 2016) and rearrayed using the S&P BioMatrix3 - BC robot to create barcoded yeast ORF-eome (byORFeome) collection.

 

Yeast strains: The initial GFP-reporter strains were engineered by integrating yeast-optimized superfolder GFP (Pédelacq et al, Nat Biotech, 2006; Thorn et al, Plos One, 2013) under the GAL1 promoter either into URA3 (RY2011-MATa and RY2031-MATα) or HO (RY3010 and RY3030) loci of the previously described RY1010 (MATa) and RY1030 (MATα) BFG Tool-kit strains (Yachie et al, MSB, 2016). RY3011 and RY3031 strains were generated by mating RY2011(MATa) and RY3030 (MATɑ) and dissecting spores. One MATa and four MATɑ strains were selected by streaking spores onto SC-URA+NAT plates. Mating type was verified by alpha factor testing (inhibits growth of MATa cells) and by mating to MMY5225(MATɑ)/mmy5228(MATa) strains (gift from Marc Meneghini lab) which are complete prototrophs.

 

Vectors details:

 

 

Y2H screens and assay versions:
All-by-all pairwise space of the yeast proteome was divided into 16 subspaces containing ~1.5K x 1.5K bait-prey combinations (represented by ~2.5K x 2.5K barcodes), where baits with similarly high background activities were grouped in the same subspace. Each baits and presy from subspace were en masse transformed into RY3011 and RY3031 strains, respectively. The successful transformants were scraped and pooled all-by-all mating were performed for each subspace ensuring an average of ~100 diploid CFUs for each protein pair combination. Then plates were scraped and frozen as glycerol stocks to create high coverage diploid pools representing each subspace. During screening glycerol stocks were diluted in 2L cultures at 1OD and were grown for 24h in baseline, with an additional 24h growth in each perturbed environments. Interactions were then identified via FACS, ensuring ~100 cells per protein pair coverage, followed by PCR amplification and deep sequencing of recombined bait-prey barcodes. Environmental Yeast Reference Interactome (EYRI) contains protein interactions that are either apparently dynamic (present in 1 to 3 out of tested 4 environments) or static (present in all 4 of the tested environments) according to our high-saturation retest (HSR) pipeline (Sheykhkarimli et al, 2024), where every hit from every environment during the primary global all-by-all screen is retested again (ie.subjected to verification) in all 4 environment. Afterwards, a representative subset of ~1K interactions were subjected to independent N2H validation along with the Positive (105 PPIs reported at least 5 times in the literature) and Random (178 protein pairs extremely unlikely to interact) Reference Set pairs. GAL1::HIS3 reporter gene.

 

Acknowledgments

CCSB interactome mapping and ORFeome cloning efforts are supported by federal grants from the National Human Genome Research Institute.