Aki Nishimura

(Bayesian) Statistician / Data Scientist for Public Health / Statistical Software Developer / Assistant Professor @jhubiostat

[Aki Nishimura @_aki_nishimura]

Postdoc opening to work on @OHDSI pharmaco-epi projects with myself and others at Johns Hopkins Center for Drug Safety and Effectiveness: https://t.co/YXm0QsRpY8

If interested, email or catch me at the OHDSI symposium this week!

[Aki Nishimura @_aki_nishimura]

Please apply/spread the word! Fwiw, I am a faculty member of the institute and would be more than happy to serve as your (co-)mentor.

[Johns Hopkins Data Science and AI Institute @HopkinsDSAI]

Join us in advancing data science and AI research! The Johns Hopkins Data Science and AI Institute Postdoctoral Fellowship Program is now accepting applications for the 2026–2027 academic year. Apply now! Deadline: Jan 23, 2026. Details and apply: https://t.co/bdxrDTzc4Q

[Aki Nishimura @_aki_nishimura]

Celebrating Yuzheng Dun's successful defense @jhubiostat (co-advised by @nilanjan10c and me) at the famed Mama Chang near DC. Joy to work with you and see you come out with an impressive PhD thesis, Yuzheng! Looking forward to seeing what you accomplish next with @wu_biostat 😉

[Aki Nishimura @_aki_nishimura]

And here's the second in the series, significantly generalizing the HMC-PDMP connection suggested by the first: https://t.co/iF7ZYE33vZ

The third is to be arXived soon and extends the connection to the "sticky" algorithms for posterior inference under spike-and-slab priors.

[Aki Nishimura @_aki_nishimura]

Our paparazzi article on the secret kinship between HMC and PDMP is now officially out! https://t.co/VfcJThZrs4

This is the first in a series of work showing PDMP to be essentially a version of (bouncy) HMC, shedding light on their relative strengths and limitations.

[Aki Nishimura @_aki_nishimura]

Congrats to Fan and other Blackwell-Rosenbluth awardees for well-deserved recognition. And thank you @j_ISBA and the award committee for making this award possible. It's so important for the @ISBA_events community to continue recognizing contributions of our junior researchers!

[Fan Bu @FannyBu214]

Very thrilled, and humbled. Thank you so much for this recognition!

[Aki Nishimura @_aki_nishimura]

US election might be occupying much of everyone's mind nowadays, but let's not forget about other important elections. Like the @ISBA_events one to elect new board members for your own communities!

(Fwiw, yours truly is running for "Bayesian Computation Secretary.")

[jhubiostat @jhubiostat]

Come join @jhubiostat! Learn about our open positions on our website. Review of applications will begin November 1: https://t.co/MpiyYuMbb4

[Aki Nishimura @_aki_nishimura]

Congrats to @rohan_khera, his team at Yale, and all the @OHDSI collaborators for delivering this landmark evidence on diabetes drugs. It's been a huge honor to be involved in the LEGEND initiative and working with y'all.

[OHDSI @OHDSI]

The LEGEND-T2DM team studied the effectiveness of several types of diabetes medications to reduce cardiovascular events across nearly 1.5 million patients. Check out the @JACCJournals study ⬇️ #JoinTheJourney @suchard_group

https://t.co/z7DzTK1XYr

[Aki Nishimura @_aki_nishimura]

Very exciting to have our paper featured in the National Eye Institute's latest challenge, as a pioneering work bringing the power of @OHDSI to ophthalmology: https://t.co/5V4EJh2TES

Congrats @ CindyCai and the team; let's continue our Heraculian efforts to save more Sisyphuses

[Aki Nishimura @_aki_nishimura]

Announcing @jhubiostat reception Mon, Aug 5th at #JSM2024! 🥂🎉 8 - 10pm @ Skybox Deck of Garden Tavern: https://t.co/K6NISTnuqX.

The venue is part of the @cartsidepdx food cart pod, so you can also grab a pre-reception dinner from food carts for an iconic Portland experience!

[Aki Nishimura @_aki_nishimura]

Looking forward to sharing my latest work at @j_ISBA's Blackwell-Rosenbluth awardee session at #jsm2024. Our *bouncy* HMC unifies HMC and PDMP, two cutting-edge paradigms in Bayesian computation: https://t.co/iF7ZYE2vGr.

Big thanks to @Bea_Franzo &
@FannyBu214 for organizing!

[Aki Nishimura @_aki_nishimura]

My first foray into polygenic risk score w/ awesome statsgen colleagues @nilanjan10c & @Jin_10ustc: https://t.co/khTYrp2Eza. Turns out the commonly-used approx likelihood can give rise to an improper posterior! (We present a fix.) It won @YuzhengDun an ASA student paper award 🎉

[Aki Nishimura @_aki_nishimura]

@emmaldry Congrats & keep it up!

[Elizabeth Stuart @Lizstuartdc]

Another job posting! This time for finance staff folks.
@jhubiostat is looking for a Sr. Grants and Contracts Analyst to provide support on pre-award and post-award functions. Join us!

https://t.co/ShyFXuiSxa

[Karl Rohe @karlrohe]

Just got an email from someone who already got a job offer with an expiring timeline. I think this is too bad.

my response:

[Aki Nishimura @_aki_nishimura]

Congrats @FannyBu214 & Hats off for your leading role in OHDSI's collaboration with FDA's CBER Biologics Effectiveness and Safety (BEST) System 😉

[OHDSI @OHDSI]

Congratulations to lead author Fan Bu and the team on the recent publication in Statistics in Medicine: Bayesian safety surveillance with adaptive bias correction. @suchard_group @kricketchirps @_aki_nishimura

https://t.co/kQrPsJh8zQ

[Richard Everitt @bayesian_stats]

Join us at @warwickstats! We are advertising for 5 Assistant Professors (3 in Statistics; 2 in Computational Statistics or Machine Learning) and one Associate Professor. Details at https://t.co/qfj6gPLlql

[Aki Nishimura @_aki_nishimura]

Glad to see my academic sister @FannyBu214 from Duke Stats/UCLA Biostats already emerging as an important figure. She must have caught the same save-the-world-with-biostats bug from @suchard_group 🙃

[ENAR @ENAR_ibs]

Please join us in welcoming Dr. Fan Bu, one of the three new members of our social media committee!

[Rohan Khera @rohan_khera]

In @BMJMedicine, we report the largest federated pharmacoepi study of 2nd line diabetes drugs in CV disease

Multinational - with 4.8 million ppl - but w/o pt data sharing across sites!

A new way to epi - details in 🧵

w @suchard_group #LEGEND Team
https://t.co/LTbmBmxL3W