collect-columns

This tool retrieves a column from each in a set of tables and compiles into a single table. Optionally, additional attributes from the associated GTF/GFF file may be added to the output tables.

Installation

Install from PyPI: pip install collect-columns

Install from github:

• Clone the repository: git clone https://github.com/biowdl/collect-columns.git
• Enter the repository: cd collect-columns
• Install using pip: pip install .

Usage

collect-columns output_path input_files...

It assumes that all input count tables are in the same format. By default the format is assumed to be headerless and tab separated, with the first column being the feature identifiers and the second the values of interest. The output table will use the same separator as the input tables and contain a header. The feature column will contain the feature identifiers, the value columns will be named after the input files or according to the names given through the -n option, which takes a list of names as argument.

Please note that if multiple rows with the same feature identifier exist in an input table, then these values will overwrite each other in the output table by default. See also the -S flag.

In order to use a different input format the following options can be given:

To add additional attributes from a GTF/GFF, the following options can be given:

Examples

HTSeq-count

Using the output from HTSeq-count as input the following command:

collect-columns all.tsv s1.tsv s2.tsv

will result in a table like:

Stringtie

Using stringtie abundance output as input, the following command:

collect-columns all.FPKM s1.abundance s2.abundance \
    -c 7 \
    -H \
    -a ref_gene_id gene_name \
    -g merged.gtf \
    -n sample1 sample2 \
    -S # Stringtie may at times return multiple rows for one gene, these values can simply be summed up.

will result in a table like: