Synthetic Protein Interaction Dataset

Dataset overview This dataset contains synthetic protein–protein interactions designed to reproduce the structure of a simplified biological interaction network. Each record represents an interaction between a source protein and a target protein, together with information about the interaction type, direction, regulatory effect, organism, and entity type. The dataset was created for software development, testing, teaching, demonstrations, and workflow validation. It is intended for situations where realistic protein-interaction data structures are needed without relying on production databases or experimentally validated interaction records. Dataset contents The dataset contains 23 protein–protein interaction records covering: 15 unique protein accessions 16 distinct gene-symbol values as written in the file Five interaction types: activation binding inhibition phosphorylation ubiquitination Three organisms: Homo sapiens, NCBI Taxonomy ID 9606 Mus musculus, NCBI Taxonomy ID 10090 Rattus norvegicus, NCBI Taxonomy ID 10116 The data are provided as a tab-separated values file: synthetic_protein_interactions.tsv Each row describes a source protein, a target protein, their corresponding gene symbols and organisms, and the properties assigned to the interaction between them. Additionally, this dataset includes a file that follows the MLCroissant specification MLCroissant specification. This file describes the metadata for this dataset. Data fields Column Data type Description Example Constraints / Notes source String Protein accession assigned to the source node of the interaction. P08228 The same protein may occur in multiple rows and may appear as either a source or target. target String Protein accession assigned to the target node of the interaction. P00533 Together with source, identifies the pair of proteins associated with the interaction. source_genesymbol String Gene symbol corresponding to the source protein. SOD1 Gene-symbol capitalization should be treated consistently during downstream processing. target_genesymbol String Gene symbol corresponding to the target protein. EGFR The same biological gene may be represented by differently capitalized strings in synthetic test cases. is_directed Integer / Boolean flag Indicates whether the interaction has a defined direction from the source protein to the target protein. 1 1 indicates a directed interaction; 0 indicates an undirected interaction. is_stimulation Integer / Boolean flag Indicates whether the interaction is assigned a stimulatory or activating effect. 0 1 indicates stimulation; 0 indicates that stimulation is not assigned. is_inhibition Integer / Boolean flag Indicates whether the interaction is assigned an inhibitory effect. 0 1 indicates inhibition; 0 indicates that inhibition is not assigned. consensus_direction Integer / Boolean flag Indicates whether the interaction direction is considered the consensus direction in the synthetic record. 1 1 indicates consensus support for the direction; 0 indicates no consensus direction. consensus_stimulation Integer / Boolean flag Indicates whether a stimulatory effect is represented as the consensus regulatory effect. 0 1 indicates consensus stimulation; 0 indicates no consensus stimulation. consensus_inhibition Integer / Boolean flag Indicates whether an inhibitory effect is represented as the consensus regulatory effect. 0 1 indicates consensus inhibition; 0 indicates no consensus inhibition. type String Synthetic interaction category assigned to the protein pair. binding Values in the current dataset include activation, binding, inhibition, phosphorylation, and ubiquitination. ncbi_tax_id_source Integer NCBI Taxonomy identifier assigned to the organism of the source protein. 10090 Values in the dataset are 9606, 10090, and 10116. entity_type_source String Type of biological entity represented by the source node. protein Constant value in the current dataset. ncbi_tax_id_target Integer NCBI Taxonomy identifier assigned to the organism of the target protein. 9606 Values in the dataset are 9606, 10090, and 10116. entity_type_target String Type of biological entity represented by the target node. protein Constant value in the current dataset. Intended uses This dataset may be useful for: testing protein-interaction data-processing pipelines prototyping protein–protein interaction networks developing biological knowledge graphs testing directed and undirected graph representations validating regulatory interaction logic testing network visualization and analysis tools developing database schemas and APIs creating software tutorials and educational examples validating data-quality, transformation, and integration procedures developing unit, integration, and regression tests Data quality The dataset contains no missing values. It contains one completely duplicated interaction record. This duplicate may be useful for testing duplicate-detection and data-cleaning procedures, but users should remove it when a unique-record dataset is required. The dataset also includes a gene-symbol capitalization inconsistency: GAPDH and gapdh occur as separate string values. This may be used to test normalization and case-sensitivity handling. The interaction records include both directed and undirected cases and use integer values, 0 and 1, to represent boolean properties. Because the data are synthetic, combinations of direction, stimulation, inhibition, and consensus flags are intended primarily for software testing and should not be interpreted as biologically validated regulatory logic. Synthetic-data statement All protein–protein interaction records in this dataset are synthetic. The dataset uses real or biologically recognizable protein accessions, gene symbols, and NCBI Taxonomy identifiers to provide a realistic data structure. However, the interaction types, regulatory effects, directionality properties, consensus annotations, and relationships between proteins were created for testing and demonstration purposes. The presence of a real protein accession or gene symbol does not imply that the corresponding synthetic interaction is biologically valid. None of the protein–protein interactions in this dataset should be interpreted as experimentally validated biological findings. The dataset is not an official export from UniProt, NCBI, Reactome, SIGNOR, IntAct, BioGRID, or any other biological interaction resource. It does not reproduce the biological accuracy, coverage, evidence model, or curation standards of those databases. Limitations This dataset should not be used for: clinical interpretation biomedical decision-making biological discovery experimental validation prediction of regulatory mechanisms protein-interaction confidence assessment pathway or network enrichment conclusions claims about real protein–protein interactions comparisons of interaction evidence across biological databases Users requiring authoritative protein-interaction information should consult official biological databases and their corresponding release and evidence documentation. File format The dataset is distributed as a UTF-8 encoded TSV file. The first row contains the column headers, and each subsequent row represents one synthetic protein–protein interaction. Boolean properties are encoded as integers: 1 represents true 0 represents false