Ruhollah (Roohy) Shemirani

Statgen at Regeneron Genetics Center, USC alumni

[Alexander G. Ioannidis @alexGioannidis]

Our ADAMIXTURE method is now out in Bioinformatics!

Scaling genetic clustering (ADMIXTURE/STRUCTURE-style unsupervised clustering) is now a solved problem imo. ADAMIXTURE can cluster all of @uk_biobank (half a million genomes) in just over an hour with likelihoods as good as ADMIXTURE, which would take weeks.

@JoanSaurina did an amazing job efficiently coding this! ADAMIXTURE is faster than any existing method and with none of their compromises. For comparison, the @AllofUsResearch paper clustering dropped to only 150k variants (far below what would be sufficient for fine scale local ancestry) then projected down to only 16 dimensions 😳 (and those were supervised PCs, so the method was no longer unsupervised, but confined to the 3k reference samples' ancestry space). Other methods make other compromises and are all still much slower.

https://t.co/juPq5OQAVm

[Genevieve Wojcik, PhD MHS @genandgenes]

📢Excited to share our comprehensive review led by @EstherHL5 on gene-environment interactions.

Within, you'll find:
🛠️Summary of current methods by underlying hypothesis
🗺️Framework for considerate epidemiological modeling
🗣️Call for greater contextual diversity needed for GxE

[Christa Caggiano @ChristaCaggiano]

Our work on fine-scale identity-by-descent clusters in the UCLA ATLAS Biobank was published online today! I'm immensely proud of this work and its implications for the health of diverse populations in my home, Los Angeles 🌴
https://t.co/TLvfxvGelR

[Icahn School of Medicine at Mount Sinai @IcahnMountSinai]

An international team of scientists has assembled the first human “pangenome” – a more representative reference genome that captures almost all the genetic variability residing in the DNA of humans around the globe. Dr. @EimearEKenny shares more: https://t.co/l7z05DKQVI

[Eimear Kenny @EimearEKenny]

Announcing the first draft of the new human 'pangenome' reference representing 47 diverse humans, ~120M new bp, >1000 new gene dups.

Congrats to @khmiga @BenedictPaten @lh3lh3 @irahall9 @tobiasmarschal @erikgarrison and all in the @HumanPangenome

https://t.co/n4gRczm7vi

[Jennifer Blanc @jblanc12]

Super excited to share a preprint of work from my PhD with @JeremyJBerg 🎉 https://t.co/AGh8syVb9d

[Jerome Kelleher @jeromekelleher]

#msprime running in the browser with genuinely respectable performance via Pyodide magic. Amazing! Early preview of the possibilities here, I'm sure this will be a killer app for teaching. Nice one Ben!💪

[Jerome Kelleher @jeromekelleher]

Hard to overstate how big a milestone this is for the #tskit project - this is the start of "Long Term Support" for the C API. You can build your C/C++/RCpp/whatever application with tskit support built-in and be sure of long-term stability. 1/3

[Genevieve Wojcik, PhD MHS @genandgenes]

Essential reading for those in human genetics research.

Outlines our ongoing responsibility to understand how research is used and twisted in white supremacist ideologies with suggested paths forward so that we can actively work against it.

🧵👇🏼

[Genevieve Wojcik, PhD MHS @genandgenes]

Human genetics has a diversity problem.

Not just in terms of study pops, but also in who asks the questions.

We urgently need diversity in our research teams re: expertise and lived experiences to better see each others blind spots and do better.

Let me explain.

🧵(1/N)

[Ruhollah (Roohy) Shemirani @roohys]

Dear PIs in genetic/genomics,
I know a brilliant young scientist from Iran looking to expand her experience before starting a Ph.D. If you are looking for an intern, please DM me for more info/CV.

[Michael Deibert @michaelcdeibert]

Mr Lordi, singer of Finnish heavy metal band Lordi, receives his second dose of Covid-19 vaccine in Rovaniemi, Finland on August 1, 2021.

[Ruhollah (Roohy) Shemirani @roohys]

@EimearEKenny @popgenepi Looking forward to doing this work for the foreseeable future!! 🙏🙏🙏🤞

[Ruhollah (Roohy) Shemirani @roohys]

Check out our paper on iLASH: our scalable IBD detection algorithm for accurate analysis of biobank-scale datasets. A close collaboration between @USC_ISI, @CUAnschutz , and @SinaiGenHealth.

[Nature Communications @NatureComms]

iLASH is an algorithm for rapid estimation of identical-by-descent genomic regions, applicable to mega-scale datasets.

@popgenepi @EimearEKenny @GillianBelbin @Christy_Avery @roohys @USC_ISI @USCViterbi @USC @SinaiGenHealth @IcahnMountSinai @CUAnschutz

https://t.co/tWGDRYWRXv

[Ruhollah (Roohy) Shemirani @roohys]

@GillianBelbin @EimearEKenny @popgenepi Glad you're back!!! You are very kind as always 🙏. I am also very proud of this 5-year project of ours!
Add to that the fact that we get to continue doing this stuff...

[Ruhollah (Roohy) Shemirani @roohys]

We also included a deep analysis of IBD networks in the PAGE study and UK Biobank datasets in the paper in which Dr. Belbin provides great population genetics insights only available through the analysis IBD nets.

[Ruhollah (Roohy) Shemirani @roohys]

All of this prallelization is automatic, you do not need to deal with complicated computing cluster configuration to benefit from it.