Barış Ekim  

@[email protected]
118 Followers
67 Following
17 Posts
🇹🇷. PhD student in algorithmic bioinformatics at MIT. Tooting in TR/EN, views my own.
Websitehttps://people.csail.mit.edu/ekim
Githubhttps://github.com/ekimb
Twitterhttps://twitter.com/ekimbarisc
Keybasehttps://keybase.io/ekimb
Barış Ekim  Dec 27, 2022
Nils HomerDec 27, 2022
Mapquik, a new read mapper for HiFi reads, is 30-35x faster than the state of the art, with minimal los in accuracy. And it has a fantastic logo to boot! @baris #bioinformatics
Code: http://github.com/ekimb/mapquik
Preprint: https://www.biorxiv.org/content/10.1101/2022.12.23.521809v2
GitHub - ekimb/mapquik: Efficient low-divergence mapping of long reads in minimizer space

Efficient low-divergence mapping of long reads in minimizer space - GitHub - ekimb/mapquik: Efficient low-divergence mapping of long reads in minimizer space

GitHub
Show threadBarış Ekim  Dec 27, 2022
@bioinformatics Thanks for the boost, Nils! :)
Barış Ekim  Dec 20, 2022

What are y'all's favorite LaTeX templates/document classes to use for a submission to a journal that doesn't provide a template these days?

They do have a couple requirements I could add on pretty easily (spacing etc.), but in general I'm just looking for something pretty.

Barış Ekim  Dec 17, 2022
The obvious answer is k = 19, since it's a useful k-mer length AND the smallest prime that's the sum of three distinct primes (not sure how that's useful - I just think it's neat)!
---
RT @gaetan_benoit
I don't know why but I suddenly want to create k=15 shirts 😈
https://twitter.com/gaetan_benoit/status/1604203773991190529
Gaetan Benoit on Twitter

“I don't know why but I suddenly want to create k=15 shirts 😈”

Twitter
Show threadBarış Ekim  Dec 8, 2022
It worked!
Barış Ekim  Dec 8, 2022

#algorithmicbioinformatics peeps!

Say, hypothetically, I wanted to test a new read mapper/pairwise aligner for Nanopore reads.

For synthetic reads, what is the length and error-rate I should aim for, and is there a specific tool to simulate Nanopore reads (I'm using pbsim with 10% and 25kb right now, and I'm aware of nanosim, though I'd like to use something simpler, if it exists)?

For real data, what's the most current read set (and reference) people like to use for benchmarking?

Thanks!

Barış Ekim  Dec 8, 2022
Testing cross-posting...
Show threadBarış Ekim  Nov 17, 2022
@jtnystrom Do you have a public repo for it?
Show threadBarış Ekim  Nov 17, 2022

@mikelove 👋

I agree with @emiller - Twitter feels more public (especially to senior academics with thousands of followers), and (from my experience), they feel reluctant to engage with trainees/grad students.

So far, my experience here has been the opposite, though - and I like that. For me, it's never been about promoting work (I have nothing to promote anyway), but I'm hoping to get insight + pointers from senior academics (which I was hoping I'd find at Twitter, and that was disappointing).

Barış Ekim  Nov 13, 2022
mariafarrellNov 13, 2022

Just a week or two in and I fking love this place so much. It's like stumbling out of the noise and into the kitchen at a party, and finding your people right there all the time.

You miss the friends you left behind in the shouty room where most who talk to you are looking over your shoulder to see if there's anyone more important to latch onto, but you hope your friends will soon find the kitchen, too.