KINference

This repository contains the source code of KINference: Data-Driven Inference of Kinase Interaction Networks alongside two example datasets.

Installation instruction

1. Install with:

devtools::install_github('bionetslab/KINference')

How to run KINference

1. Input provided as intensity matrices

library(Kinference)

run_KINference(
    x0.path = 'data-raw/example_data/example_x0.tsv',
    x1.path = 'data-raw/example_data/example_x1.tsv',
    apply.CORR = TRUE,
    m = 9,
    output.id = 'example_matrix_run', 
    output.path = 'example_results/'
)

2. Input provided in vector format

library(Kinference)

run_KINference(
    f.path = 'data-raw/example_data/example_f.tsv',
    output.id = 'example_vector_run', 
    output.path = 'example_results/'
)

3. Only computing kinase enrichments

If you only want to compute kinase enrichments, set all filters to FALSE: apply.DIFF = FALSE, apply.FS = FALSE, apply.PCST = FALSE, and apply.CORR = FALSE.

Input specification

KINference can be run with either 2 input matrices of intensity measurements of phosphorylation sites or 1 input vector of log2FC-transformed intensities. The inputs have to be provided as tab-separated files. The matrices and the vector must contain one column named Protein. This column contains the annotations of the UniprotIDs and the phosphorylated amino acid (AA) and its sequence position (Pos) in UniprotID_AAPos format. See the example datasets in data-raw/example-data/ for reference.

Parameters of KINference

• x0.path: Path to the first intensity matrix file. Default is NA.
• x1.path: Path to the second intensity matrix file. Default is NA.
• f.path: Path to the intensity vector file. Default is NA.
• output.path: Directory where the results will be saved. Default is 'results'.
• output.id: Identifier for the output files. Default is 'key'.
• species: Species name, either 'Homo sapiens' or 'Mus musculus'. Default is 'Homo sapiens'.
• translate_uniprots: If set to false, the Uniprot IDs will not be translated to human Uniprot IDs. If set to FALSE, the PCST filter will be automatically disabled because the PCST filter needs a connected network and, therefore, needs all Uniprot IDs to be human Uniprot IDs because the provided kinase data is only for human kinases. Default is TRUE.
• paired.samples: Logical indicating if the samples are paired (see paper for difference in paired vs unpaired computation of log2fc). Default is TRUE.
• apply.log2: Logical indicating if log2 transformation should be applied to the data. Default is FALSE.
• n: Number of top kinases to infer. Default is 15.
• alpha: Parameter for baseline kinase inference. Default is 0.9.
• apply.DIFF: Logical indicating if the DIFF filter should be applied. Default is TRUE.
• apply.FS: Logical indicating if the FS filter should be applied. Default is TRUE.
• apply.CORR: Logical indicating if the CORR filter should be applied. Default is FALSE.
• apply.PCST: Logical indicating if the PCST filter should be applied. Default is TRUE.
• beta: Parameter for node filter computation. Default is 0.4.
• gamma: Parameter for kinase enrichment computation. Default is 1.0.
• delta: Parameter for edge filter computation. Default is 0.8.
• epsilon: CORR significance threshold. Default is 0.05.
• m: CORR minimum sample threshold. Default is 10.
• multiple_testing_correction: Method for multiple testing correction for CORR filter. Default is 'BH'.
• custom_serine_threonine_kinase_data.path: Path to a custom serine/threonine kinase data file. Default is NULL.
• custom_tyrosine_kinase_data.path: Path to a custom tyrosine kinase data file. Default is NULL.