Pejvak Moghimi

Applied deep learning research scientist. PhD: using DL for predicting convergence across large-scale immune repertoire genomics data.

[iReceptor Gateway @ireceptorgw]

@thesugarscience @Bruskolab @ufdiabetes @BarbaraDavisCtr @UMassDiabetes @T1D_Research @BRIseattle @RepertoireIM @AdaptiveBiotech @scrippsresearch @felixbreden @AMP_CMDKP How about a grand challenge in #FAIR sharing of #AIRRseq #BCell/#TCell data so we have enough data to train ML models. A necessary first step in the grand challenge of using ML models to predict #TCR antigen specificity... Join us in building the @airr_community #AIRRDataCommons.

[Nature Biotechnology @NatureBiotech]

Single-sequence protein structure prediction using a language model and deep learning https://t.co/xlHxIEqQkt

[Rahul Satija @satijalab]

Stunning preprint from @WangXiaoLab and @ganoopyliujia introducing STARmap PLUS for in-situ transcriptome profiling. Used to generate a 3D map of the whole mouse brain and spinal cord (1,000 genes, 1M+ cells)
https://t.co/l7XBQuqwGR

[Glenn Greenwald @ggreenwald]

This is, without exaggeration, one of the most amazing things I've ever seen on television, and tells you all you need to know about the state of mainstream US discourse and the implicit premises on which it's based:

[GreiffLab 💻🔬💊 @victorgreiff]

Over the moon that our study on the predictability of antibody-antigen binding is featured on the cover page of @CellReports! The cover was designed by our resident scientific illustrator @pandaisikit (who's also the first author).
Link to paper: https://t.co/736WAtjXLy

[Pejvak Moghimi @IranianITcrowd]

This sounds incredible for anyone interested in phylogenetics

[(Fred)erick Matsen @ematsen]

I'm happy to report a strong score on a grant application to develop variational Bayes phylogenetic analysis, joint with @suchard_group @m4ment and Cheng Zhang.

It's too early to make a formal job ad, but we'd like to hear from any interested postdoc candidates.

🧵 about goals:

[Pejvak Moghimi @IranianITcrowd]

This nice quote always hits very close to home for me, but particularly right now.

[Ewan Birney @ewanbirney]

Many people think science is about getting answers; it is actually as much about working out the right questions to ask.

[Pejvak Moghimi @IranianITcrowd]

I know quite a few people who could take note.

[Prof. Feynman @ProfFeynman]

If you cannot explain something in simple terms, you don't understand it yourself.

[Pejvak Moghimi @IranianITcrowd]

@GeorginaGirt Lol. This would had been an improvement on the episode, or, the whole season.

[Pejvak Moghimi @IranianITcrowd]

@Philow52348876 @chama_tahiri @zackbeauchamp @AidaAlami After all the nonesense about Andals, someone finally said it, or I would have pulled my hairs out. 2010s has certainly been the decade of "failed smartassery".

[Mohammed AlQuraishi @MoAlQuraishi]

From Google AI this time.

[Jessica Schleider, PhD @JSchleiderPhD]

Just learned that I received a grant for which my cat accidentally clicked 'submit' while I was out walking the dog

Mochi now has a higher grant hit-rate than I do and will be submitting all of my applications until further notice

Here she is looking unimpressed:

[Soumith Chintala @soumithchintala]

Pretty excited about NVIDIA's Jetson Nano. 5W, $99 and a 128-core Maxwell sounds pretty great. This year is so good for embedded deep learning!
https://t.co/34EMapE9Sw

[Pejvak Moghimi @IranianITcrowd]

@skleinstein Hands on session is a great idea!
1. Perhaps, an independent webinar for each immcan tool, specially phylogenetics?
2. A session on repertoire annotation would be great; there's so much hidden jargon and concepts on the seq biojnformatics but little proper documentation.

[Steven Kleinstein 🖥 @skleinstein]

Want to analyze B cell receptor repertoire sequencing data, but not sure how to get started? Register for our webinar: "Introduction to B cell repertoire analysis using the https://t.co/tHinF6zRsL Framework” on 3/22 at noon ET: https://t.co/Hwn44NUbDs. #AIRRseq #RepSeq #BCR

[Hugh Kearns @ithinkwellHugh]

The literature is now so vast that there isn't enough time to read every paper in depth. You need to learn how to scan or filter papers. Here's a three pass approach.

[Prof. Feynman @ProfFeynman]

The FEYNMAN technique of learning:

STEP 1 - Pick and study a topic 📖

STEP 2 - Explain the topic to someone, like a child, who is unfamiliar with the topic 🗣️

STEP 3 - Identify any gaps in your understanding 🔍

STEP 4 - Review and Simplify! ♻️