Design the Cassandra database schema for variants

[schae234]

I need to know if the philosophy behind the Cassandra schema differs from the one we used for SQLite. I know that Cassandra believes in denormalization so we might be sacrificing efficiency for space (which might have sped up SQLite in the first place)

[schae234]

Query Planning

• Q1*. Find all chromosome names (from locuspocus)
• Q2*. Find all SNPs on Chip (from locuspocus)
• Q3*. Find all individual names (from minus80)
• Q4. Find the dosage for all SNPs on chrXXX for [individuals]