search

amplab / snap

Scalable Nucleotide Alignment Program -- a fast and accurate read aligner for high-throughput sequencing data
https://www.microsoft.com/en-us/research/project/snap/
Apache License 2.0
287 stars 66 forks source link

Parameter tuning? #34

Closed ryananeff closed 9 years ago

ryananeff commented 9 years ago

Hi Ravi and the SNAP team,

Very interesting and excellent talk at ASHG. Do you have any information about parameter sweeps for other flags than the seed length (i.e. ROC curves for MaxDist, MaxHits, etc.) to use for tuning against a well-characterized dataset?

-Ryan N.

bolosky commented 9 years ago

It’s been a while since we’ve done it, but other than seed length the only parameter that has a really major effect on alignment quality is maxHits. It also, unsurprisingly, has a big effect on speed which is inverse to quality.

For the paired-end aligner, there’s a decently large speed effect from numSeeds, but it doesn’t seem to affect quality much, so we left it low by default. It doesn’t matter nearly as much for single end.

--Bill

From: Ryan Neff [mailto:notifications@github.com] Sent: Tuesday, October 21, 2014 10:07 AM To: amplab/snap Subject: [snap] Parameter tuning? (#34)

Hi Ravi and the SNAP team,

Very interesting and excellent talk at ASHG. Do you have any information about parameter sweeps for other flags than the seed length (i.e. ROC curves for MaxDist, MaxHits, etc.) to use for tuning against a well-characterized dataset?

-Ryan N.

— Reply to this email directly or view it on GitHubhttps://github.com/amplab/snap/issues/34.