PICKLE: A human protein-protein interaction meta-database
PICKLE (Protein InteraCtion KnowLedgebasE) is a meta-database for the human direct protein-protein interactome, integrating publicly available source protein-protein interaction (PPI) databases via genetic information ontology.
PICKLE integrates the primary PPI datasets by superimposing them on the UniProtKB/Swiss-Prot reviewed human complete proteome (RHCP) ontology network without any a priori transformations.
This is a major novelty over the most commonly used methodology of primary PPI dataset integration after their normalization to a pre-selected level of genetic reference. Importantly, the ontological PPI dataset integration allows the
PICKLE heterogeneous integrated network to be reversibly normalized to any level of genetic reference without any loss of the original PPI information, the latter being used for the identification of normalization biases. Moreover, it enables the appraisal of potential false positive interactions through PPI source database cross-checking.
The PICKLE interface allows for the simultaneous query of multiple entities and provides integrated human PPI networks at either the protein (UniProt) or the gene level, at three PPI filtering modes: unfiltered, standard and cross-checked (default), using a standardized
confidence scoring protocol for PPIs being direct, based on the quality of the supporting experimental evidence.
PICKLE Release 2.6 (Oct 2 2020)
The UniProt-normalized default PICKLE network contains:
- UniProt IDs: 16502
- Interactions: 200845
- Supporting Publications: 43670
The Gene-normalized default PICKLE network contains:
- Entrez Gene IDs: 16333
- Interactions: 201874
- Supporting Publications: 43614