Hi,
I am using ld_prune() after performing some variant filtering with the following command:
pruned_variant_table = hl.ld_prune(scz_varqc.GT, r2=0.2, bp_window_size=500000)
scz_varqc_pruned = scz_varqc.filter_rows(hl.is_defined(pruned_variant_table[scz_varqc.row_key]))
Response so far:
[Stage 3:=====================================================> (124 + 3) / 127]2020-02-17 13:56:23 Hail: INFO: ld_prune: running local pruning stage with max queue size of 21572 variants
[Stage 8:==================================================>(16443 + 5) / 16448]2020-02-17 15:03:21 Hail: INFO: Coerced sorted dataset
[Stage 23:=================================================>(16446 + 2) / 16448]2020-02-18 08:43:32 Hail: INFO: wrote table with 1178530 rows in 16448 partitions to file:/tmp/hail.xvD6jqsyxN8t/vjAO0Iow8b
[Stage 28:=================================================>(16445 + 3) / 16448]2020-02-18 09:50:50 Hail: INFO: Coerced sorted dataset
I get the following error:
FatalError: HailException: Cannot create BlockMatrix: filtered entry at row 0 and col 1
I am not sure why. I am running hail locally with version 0.2.24-9cd88d97bedd. Can you help please. Thank you!