Analysis optimisation

[davmlaw]

Go and collect say, examples of the slowest queries on vg.com

Then recreate them, and see if we can improve them.

It's possible that we could do some things to optimise queries, eg to make an index for low / NULL gnomAD:

CREATE INDEX idx_gnomad_af_null_or_low ON annotation_variantannotation_version_10 (gnomad_af) WHERE (gnomad_af <= 0.01 OR gnomad_af IS NULL);

Apparently you can do this in Django via:

from django.db import models
from django.contrib.postgres.indexes import Index

class MyModel(models.Model):
    gnomad_af = models.FloatField(null=True)

    class Meta:
        indexes = [
            Index(
                fields=['gnomad_af'],
                name='gnomad_af_null_or_low_idx',
                condition=models.Q(gnomad_af__lte=0.05) | models.Q(gnomad_af__isnull=True),
            ),
        ]

The other thing to look into is the sample regex query. It might be better to substring out and do string matches