fplaza / countmatrixtransformer

Compute relative gene abundance from raw read counts

count-matrix-transformer

This tool takes a raw count matrix and compute relative gene abundance according to the method exposed in Qin, J. et al. A metagenome-wide association study of gut microbiota in type 2 diabetes. Nature

The reverse operation consisting of computing a raw matrix from a normalized count matrix can also be performed.

Requirements

• A C++ compiler
• Boost C++ Libraries

Usage

./count-matrix-transformer -c <input_count_matrix> -l <genes_length_file> -t <transformation_type> -o <transformed_count_matrix> 

• -c <input_count_matrix> count matrix to transform
• -l <genes_length_file> file (without an header) which contains line by line, tab separated pairs of values '<gene name> <gene length>'
• -t <transformation_type> transformation to apply to the count matrix ('normalize' or 'denormalize')
• -o <transformed_count_matrix> output transformed count matrix

Input count matrix example

gene_id sample_1 sample_2 sample 3
gene_1 10 0 3
gene_2 0 21 8
gene_3 12 9 3
gene_4 10 12 11

genes length file example

gene_1 699
gene_2 135
gene_3 1422
gene_4 1827
gene_5 540
gene_6 61