Luka Culibrk

Bioinformatician @nyugrossman @perlmutter_cc with @MPetljak. I like climbing pointy things and clicking on virtual polygons for fun.

[Luka Culibrk @luka_culibrk]

@FSeersholm If there is a sufficient amount of DNA you could also use Duplex sequencing to perform a kind of saturation mutagenesis experiment by sequencing thousands and thousands of single DNA molecules in the sample, like in https://t.co/udHs7VXJG7 to get better estimates of selection

[Luka Culibrk @luka_culibrk]

@FSeersholm I'm guessing because the rate of mutation was so low that you couldn't do dN/dS analysis of the plasmid... What about mutational burdens of the canonical genome vs plasmid? (1/2)

[Luka Culibrk @luka_culibrk]

@rrbelloff @YounisJoseph Do you know that residency and its infamously soul crushing working schedule were developed and designed by a literal cocaine addict

[Luka Culibrk @luka_culibrk]

@tangming2005 You can also do SNP genotyping. The Sanger has a table of ~1000 cell lines' genotypes at 90-something loci. Not hard to genotype those SNPs in a sample of interest and find the true line of origin

[Lior Pachter @lpachter]

I am a computational biologist and I rewrite RNA-seq pipelines.

[Luka Culibrk @luka_culibrk]

@laxmog @LinkofSunshine welcome back, Victoria II

[Luka Culibrk @luka_culibrk]

@CroatAmerican @NYCTSubway Hell yeah brother 😭

[Luka Culibrk @luka_culibrk]

@losslandscape @deedydas Most impactful scientific research is done by trainees e.g. PhD students

Discoveries made in a lab belong to the lab (not the student)

The US' scientific prestige is built on the wealth of intelligent trainees that come to the US and make important discoveries. Most stay too.

[Luka Culibrk @luka_culibrk]

@brianwut @NathanJRobinson not in Astoria

[Luka Culibrk @luka_culibrk]

@RndmForestRunnr @garfbradaz 🤔

[Luka Culibrk @luka_culibrk]

@tangming2005

[Ming "Tommy" Tang @tangming2005]

1/ Bioinformatics takes years to master.
Not because it’s hard.
But because so much of what matters… no one writes down.
Let me explain

[Luka Culibrk @luka_culibrk]

@Phil_A_Richmond I know Mutect2 (and probably HaplotypeCaller) has that information in the GT field (when the genotype has a | instead of /, like 1|0 as opposed to 1/0), but if phasing isn't explict like that in the VCF I don't know if it's possible without going into the alignments

[Luka Culibrk @luka_culibrk]

@ladydatadoc @WallStreetApes The subway is $2.90 and faster than driving

[Luka Culibrk @luka_culibrk]

@Phil_A_Richmond @DrAshM2 Moderation. Bluesky has plenty from what I've seen and the bot situation is infinitely better than here.

[Luka Culibrk @luka_culibrk]

@johanneskoester Everyone went to Bluesky

[Luka Culibrk @luka_culibrk]

*defeated breath*
*eye roll*

[Luka Culibrk @luka_culibrk]

@jomarnz I _did_ know about hex notation in fairness, just for some reason my brain short circuited every time I would have needed to translate it to/from decimal on https://t.co/TahT35td0f

[Luka Culibrk @luka_culibrk]

SAM flags are confusing me:

samtools view -F 0x900 shouldn't output unmapped reads (because it includes 0x4), but the resulting file has alignments with 0x4

Even weirder, 0x900 includes read2, but 0x896 excludes these, despite only changing the unmapped flag

Is 0x900 special?

[Luka Culibrk @luka_culibrk]

@jomarnz Wow. What an incredibly stupid error on my part. Thanks John