You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
First, let me say I'm a huge fan of your work.
And apologies in advance, but I have a feature request-- namely you've hard-capped the minimum --max-depth to 10, and I was hoping you could take the guardrails off and drop that all the way to 1.
As for why, for those of us working w/ challenged samples it's quite easy to get bacterial contamination, which presents a lot like a PCR duplicate (but will not be marked as such). A very simple mitigation strategy is to reduce the --max-depth to something quite small; in the case of our <1x genomes, a max-depth of 1 may provide a reasonable tradeoff between the amount of data considered and its correctness. And in general, I can see taking an upper-bound from a Poisson if one wanted to set this filter dynamically to something sane (and perhaps <10).
Regardless, thanks for your help and dedication.
-August
(FWIW, I'm using the static binaries that you provide; thank you for that, btw! The version numbers are:
GLIMPSE2_phase v2.0.0 / commit = 3bed6d9 / release = 2022-12-07
)
The text was updated successfully, but these errors were encountered:
First, let me say I'm a huge fan of your work.
And apologies in advance, but I have a feature request-- namely you've hard-capped the minimum --max-depth to 10, and I was hoping you could take the guardrails off and drop that all the way to 1.
As for why, for those of us working w/ challenged samples it's quite easy to get bacterial contamination, which presents a lot like a PCR duplicate (but will not be marked as such). A very simple mitigation strategy is to reduce the --max-depth to something quite small; in the case of our <1x genomes, a max-depth of 1 may provide a reasonable tradeoff between the amount of data considered and its correctness. And in general, I can see taking an upper-bound from a Poisson if one wanted to set this filter dynamically to something sane (and perhaps <10).
Regardless, thanks for your help and dedication.
-August
(FWIW, I'm using the static binaries that you provide; thank you for that, btw! The version numbers are:
GLIMPSE2_phase v2.0.0 / commit = 3bed6d9 / release = 2022-12-07
)
The text was updated successfully, but these errors were encountered: