Bo Xia (@boxia7) 's Twitter Profile
Bo Xia

@boxia7

Junior Fellow @Harvard Society of Fellows | @GRO_Broad Fellow & PI @broadinstitute. Studying genome: how we develop, what causes diseases, what makes us human

ID: 974102729994878976

linkhttps://www.boxialab.org calendar_today15-03-2018 01:59:23

501 Tweet

2,2K Followers

1,1K Following

Bo Xia (@boxia7) 's Twitter Profile Photo

What a surprising discovery of 5-formylcytosine (5fC) — not 5hmC or 5caC — functioning as a unique activating epigenetic mark during zygotic genome activation, particularly the pol-III gene (e.g. tRNA) activation!!

Yi Ma (@yimatweets) 's Twitter Profile Photo

9.11 is still larger than 9.8, despite can memorize solutions to PhD level questions. Again, memorizing is not understanding and knowledge is not intelligence.

Anders Sejr Hansen (@anders_s_hansen) 's Twitter Profile Photo

(1/n) From Viraat Goel Ed Banigan et al. collab w blobellab Leonid Mirny biorxiv.org/content/10.110… RCMC at the M-to-G1 transition reveals strong microcompartments between CREs (E,P) in mitosis, that weaken upon G1 entry Sims reveal chromosome compaction as key regulator

Anshul Kundaje (anshulkundaje@bluesky) (@anshulkundaje) 's Twitter Profile Photo

biorxiv.org/content/10.110… Nice benchmark of single cell "foundation models" (scGPT, scFoundation) and GEARS (a GNN model) further hyped as "virtual cell models" against linear baselines on perturbation prediction. Long-story short: they can't beat the linear baselines. 1/

Anshul Kundaje (anshulkundaje@bluesky) (@anshulkundaje) 's Twitter Profile Photo

IMHO, significant progress will happen when the biological problem statement really fits the ML model design & optimization with lots of biologically informed regularization. It is extremely unlikely that "scaling" with a giant "universal learning engine" will just work. 11/11

Bo Xia (@boxia7) 's Twitter Profile Photo

Whatever way we describe it, a cell is a cell. The core is how a cell type is determined. The Weintraub experiment showed that a single TF can determine muscle cell, & Yamanaka experiment revealed OSKM (or OSN) for pluripotency. Can we find such a program for every cell type?

Lasker Foundation (@laskerfdn) 's Twitter Profile Photo

“My message to the young people is this: if I can get so close to the stars, you can too.” —Zhijian “James” Chen, 2024 #LaskerAward winner for his discovery of cGAS, an enzyme that senses self and foreign DNA #Lasker2024 #LaskerLaureate UT Southwestern Medical Center

Bo Xia (@boxia7) 's Twitter Profile Photo

Congratulations Rongxin! What an exciting time to bridge genomic research to neural circuits. Looking forward to the many new discoveries in your lab!

Weintraub Graduate Student Award (@weintraubaward) 's Twitter Profile Photo

Nominations for the 2025 Weintraub Award are now open! Celebrate the achievements of an outstanding graduate student in biology by nominating them here: fredhutch.org/en/research/di…

Itai Yanai (@itaiyanai) 's Twitter Profile Photo

-If one focuses only on what's known in one's field, then the next experiments are likely underway. What is novel is if one imports an idea from another field that others had not thought about. Top things to learn from Bob Weinberg on The Night Science Podcast 🧵⬇️ podcasts.apple.com/us/podcast/nig…