ymcki

So what is the allele frequency cutoff to call 0/0 instead of 0/1?

My understanding is that all DUPs called by sniffles are tandem repeats that are exactly one repeat more than the reference. As long as the size of the tandem repeat...

> Sorry no. so that file is just so the merge of SV candidates (internally) are more relaxed in tandem repeat regions. > > Tandem repeat are not duplications! Sniffles...

> A fix will be in the next version (v1.0.6) > > To double confirm the correct calls should be: chr1 4620711 . AAAAT A 6.41 PASS 0/1 chr1 4620715...

Speed at ~80% is better than expected. We are getting some L40Ses soon and will do the similar benchmarking. Is it possible that you also align HG002 reads to HG002v1.0.1...

My run time for HG002 0429 with sup model dataset is 12hr on 4xL40S and 8hr on 4xA100. For hac model, it is 2.5hr and 2hr. So for hac, it...

Maybe the parents overlap only at a fraction of themselves? ``` --------------------------- parent1 -------------------------------- parent2 ------------------------------------------------ duplex ```

> Hi @ymcki, > > What is happening here is most likely that you are seeing the effect of [read splitting](https://github.com/nanoporetech/dorado/commit/8ae95e732310400016f34b8f9499f1fdb4a0dd41) which was enabled in Dorado v0.4.0. In Dorado v0.3.4...

As a continuation of my benchmarking work, I looked into the alignments of the longest reads. Typically, I found that the so-called the "longest" reads (ie >200Kbp) are either a...

--basecaller_cfg '[email protected]' --remora_cfg '[email protected]_5mCG_5hmCG@v2' Kit is LSK114 4kHz