MCL_tuning: MCL Hyperparameters Tuning

In MACP: Macromolecular Assemblies from Co-Elution Profile (MACP)

MCL Hyperparameters Tuning

Description

This function optimize the choice of MCL algorithm parameter (inflation) by comparing clustering-derived partitions for each paramter values to known labels (i.e., CORUM complexes) and assess the similarity between them using quality measures including overlap score, sensitivity (Sn), clustering-wise positive predictive value (PPV), geometric accuracy (Acc), and maximum matching raio (MMR). It is recommended to first reduce redundancy in the known reference complexes via EliminateCpxRedundance, then performs parameter tuning.

Usage

MCL_tuning(hc_ppi, predcpx, refcpx, inflation = c(6, 8, 9), csize = 2)

Arguments

hc_ppi	Interactions data containing id1-id2-weight triplets.
predcpx	A data.frame containing predicted modules resulted from get_clusters.
refcpx	A list containing reference complexes (i.e., corum complexes).
inflation	A vector of integer, representing MCL inflation parameter
csize	An integer, the minimum size of the predicted complexes. Defaults to 2.

Details

MCL_tuning

Value

A data.frame containing clustering performance across different inflation values.

Author(s)

Matineh Rahmatbakhsh, [e-mail]

MACP documentation built on March 7, 2023, 7:42 p.m.