cNMTF is a data fusion framework for prioritising reliable associations between single nucleotide variants (SNVs), genes and diseases. In this webpage you will find the step-by-step process to run the algorithm on your own data using R scripts.

## Installation

Install the cnmtf package directly from the github repository. This process requires the devtools package :

install.packages("devtools")

devtools::install_github("lgl15/cnmtf")

You can check the development version of this R package on Github.

## What is inside the package?

The cnmtf package provides four categories of functions for preprocessing data, clustering, scoring SNVs and comparing results.

• Preprocessing functions:
These functions will help you to create the inputs for the algorithm.

• Factorisation functions:
Main functions to integrate the input data, generate the low-dimensional matrices and find consensus clusters.

• Scoring functions:
A set of functions to score SNVs and prioritise significant SNV-trait associations from the low-dimensional matrices.

• Comparing functions:
Auxiliary functions to compare your results across different settings of the algorithm.

## Citation

Leal, L.G., David, A., Jarvelin, M.-R., Sebert, S., Männikkö, M., Karhunen, V., Seaby, E., Hoggart, C., and Sternberg, M.J.E. (2019). Identification of disease-associated loci using machine learning for genotype and network data integration. Bioinformatics, btz310 (https://doi.org/10.1093/bioinformatics/btz310)

Corresponding author:
Luis G. Leal, lgl15@imperial.ac.uk
Centre for Integrative Systems Biology and Bioinfomatics (CISBIO)
Department of Life Sciences
Imperial College London, London, UK, SW7 2AZ

## Hands on

You must start by preparing Your input data.