Parsing GENCODE

Install the relevant packages in the usual way.

using Pkg
Pkg.add(["GeneticsMakie", "CSV", "DataFrames", "Arrow"])

To plot genes and isoforms, we need a transcriptome annotation. We can use the latest GENCODE annotation for the human genome (GRCh37), where we download the comprehensive gene annotation file in GTF format. We recommend having at least 16 GB RAM available for loading GENCODE annotation.

using GeneticsMakie, CSV, DataFrames, Arrow, Downloads
isdir("data") || mkdir("data")
url = "https://ftp.ebi.ac.uk/pub/databases/gencode/Gencode_human/release_39/GRCh37_mapping/gencode.v39lift37.annotation.gtf.gz"
file = basename(url)
isdir("data/gencode") || mkdir("data/gencode")
isfile("data/gencode/$(file)") || Downloads.download(url, "data/gencode/$(file)")
h = ["seqnames", "source", "feature", "start", "end", "score", "strand", "phase", "info"]
gencode = CSV.read("data/gencode/$(file)", DataFrame; delim = "\t", comment = "#", header = h)

The ninth column of a GTF file contains rich information about features, so we can parse this column.

GeneticsMakie.parsegtf!(gencode)

To reduce memory intake, we can also subset gencode to most commonly used columns in downstream analyses.

select!(gencode, :seqnames, :feature, :start, :end, :strand, :gene_id, :gene_name, :gene_type, :transcript_id)

To further reduce memory intake, we can instead store and load GENCODE annotation as an Arrow file.

Arrow.write("data/gencode/$(splitext(file)[1]).arrow", gencode)
gencode = Arrow.Table("data/gencode/$(splitext(file)[1]).arrow")|> DataFrame

Other transcriptome annotations, such as one from RefSeq, can be used for plotting functions as long as they contain the above columns with the right column names.

Once gencode is ready, we can look up where a gene is on the human genome.

GeneticsMakie.findgene("RBFOX1", gencode)
GeneticsMakie.findgene("ENSG00000078328", gencode)