Extracting more attributes

If you followed the previous sections, you'll have a working read_gff() and you'll have run the test and seen the important message:

++ test_read_gff(): Congratulations, all tests passed!

Congratulations!

However - there's more important information in that attributes column which we should get.

Here are a couple of challenge questions for you to try if you feel like it:

Challenge 1: extract more attributes

Challenge: extract more attributes (easy version)

In the GENCODE data, the gene_type and gene_name columns are useful as well. Update your function so it extracts these attributes into new columns too.

Challenge: extract more attributes (harder version)

Update your function so it takes an extra argument called extra_attributes, containing a list of other attributes to extract. You'd call it something like this:

In R

X = read_gff(
    "gencode.v41.annotation.head.gff3",
    extra_attributes = c( "gene_type", "gene_name" )
)

In python

X = read_gff(
    "gencode.v41.annotation.head.gff3",
    extra_attributes = [ "gene_type", "gene_name" ]
)

The function should extract these extra attributes into seperate columns as well.

Challenge 2: shrink attributes

Currently, read_gff() extracts ID and Parent (and, if you do challenge 1, other attributes too). But it also leaves these fields in the attributes column. Since the files are so big, this can waste a lot of memory.

Challenge 2: shrink attributes

Find a way to remove the extracted fields from attributes when you extract them. (Make sure to remove the semicolon too, if it's there!)

Here are some hints:

Hint 1 In R, the str_remove() function from stringr can help with this - you will need to use the right regular expression.

Hint 2 The regular expression syntax '(;|$)' will match either a semicolon or the end of the string - may be useful.

The finished product

If you get through all (or any) of that - congratulations!

For a tutorial that uses this code - see the follow-on Analysing genes across species tutorial.

Note

If instead you'd like to see my version of the finished product - the R package and the command-line program with the improvements listed here - there's also a link on the Analysing genes across species tutorial page.