Erik Garrison

``` vw -d ~/NA12878_V2.5_Robot_1.1.hhga.gz --binary --passes 1 -q ha -c -b 18 -f ~/b18passes1_robot1.model --compressed ```

Trained models on full robot1 from garvan. Tested against the second garvan set (robot2). ``` 0.002364 twostep25.model time vw -d ~/NA12878_V2.5_Robot_1.1.hhga.gz --binary --passes 20 -q ha -c -b 25 --ngram...

We are not estimating allele frequencies. We estimating genotype/site/allele truthiness. But we should be going forward. That should be pretty easy for vw because a linear model will usually suffice....

We found a problem with the previous examples. I was failing to swap out the alignments for each sample, which resulted in overfitting. These are the first results after having...

edyeet and minimap2 are fundamentally different algorithms. Both produce semantically equivalent PAF, but they get there with different approaches. minimap2 starts with minimizer chaining. Chains are built and filtered using...

Just a second note to keep things organized. I'm still learning how to use edyeet for this application, and I'm finding that a combination of large segment sizes and permissive...

> One aspect about a pangenome graph with VG that is attractive is being able to produce variant calls directly from the graph using vg deconstruct. That's the hope. Please...

You might be interested in [the pangenome graph builder, pggb](https://github.com/pangenome/pggb). It pulls together the steps for pangenome graph construction in one script.

On the website?

Ohh, word. Sorry. On Thu, Mar 12, 2015 at 1:22 PM, Henrik Feldt [email protected] wrote: > In npm, I was thinking > > — > Reply to this email directly...