Excited to announce I will start as an Assistant Professor
@Yale
@YaleMed
on Jan 1, 2024! Our lab's mission is to build tools to study cellular states in their native tissue contexts, focusing on drug response, aging, and cardiovascular disease.
Profiling the molecular heterogeneity in diverse patients is key to precision medicine & therapeutics. But single cell methods need fresh/frozen tissues (hard to scale), while FFPE tissues are routinely collected. We share single nucleus RNA-Seq for FFPE:
Excited to share our
#preprint
on inCITE-seq for measuring nuclear protein levels + transcriptome at single cell! Like CITE-seq for cell-surface proteins, we use DNA-conjugated antibodies to measure proteins, but now inside the nucleus + transcriptome.
inCITE-seq is now online at
@naturemethods
! We now include joint embedding of RNA profiles from inCITE-seq and snRNA-seq that helps resolve low RNA complexity due to fixation 1/n
Excited to share our
#preprint
on inCITE-seq for measuring nuclear protein levels + transcriptome at single cell! Like CITE-seq for cell-surface proteins, we use DNA-conjugated antibodies to measure proteins, but now inside the nucleus + transcriptome.
After 3 months of hard work, excited to unveil the Chung lab’s computational space
@YaleCVRC
@YaleMed
! Gratifying to see our vision come to life. We are recruiting both computational and wet lab postdocs - join us!
Meanwhile, many postdocs + grad students like myself would volunteer to vaccinate people 24/7 in a heartbeat! I'm dextrous and work with needles, plus also (unfortunately?) used to working long and odd hours.
That we only vaccinated 292,552 Americans today, having been able to hit 800,000 Thursday (7 Jan), speaks to the lack of any 24/7 plan or ability to execute.
Emergencies don't take off weekends.
#SingleCell
twitter: What’s the current state of the art for integrating scRNA and snRNA data from the same tissue? Each dataset is fully annotated, and most cell types are found in both, but a few types are unique to each modality. Tried Harmony and scVI so far.
On being an interdisciplinary scientist: words of wisdom from my mentor Aviv Regev on her non-traditional path, with minor contributions from me. Reported by the amazing
@metricausa
Patients on ventilators often suffer from bacterial infections. In a prospective study, we show that antibiotic resistance muts expand or contract in vivo within days of therapy during acute infection, highlighting the impt of short-term therapy choice.
New from Regev Group: multiplex samples of single nuclei for droplet sequencing with barcoded antibodies -> say bye to batch effects and save 5x $:
@broadinstitute
I’ve been waiting to share this groundbreaking work from
@AndyRusss
@jacksonweir4
and team. This is true single cell spatial multi-omics. No more pixelated spatial data. And harness all the omics built on droplet based platforms. Can’t wait to use and build with it!
Single-cell or spatial?
Our new technology - Slide-tags - allows both in the same experiment, enabling true single-cell multi-modal spatial genomics
➡️
You’ve had sushi, but have you had Korean kimbap? Filled with warm, cooked things. My simple version with garlic fried shiitake, fragrant sesame leaves, crispy pickled daikon
It was a pleasure to present inCITE-seq at Single Cell Proteomics towards bridging sc genomics and proteomics. And nothing beats the in-person discussions that followed. Thank you
@slavov_n
for the invitation and for putting together a fantastic conference!
Simultaneous single cell measurements of intranuclear proteins and gene expression
A presentation by Dr. Hattie Chung (
@hattaca
) at the 4th single-cell proteomics conference
#SCP2021
.
Cell division never ceases to amaze. It’s like an earthquake — everything about it is destructive and catastrophic, yet it’s a basic part of eukaryotic life that happens over and over again. Maybe that’s why life is a continuous cycle of growing pains?
fun to share our short work on multi-view representation of molecules
@lmrl_bio
#neurips2020
! v. proud of my high school student Jonathan Yin who led this work.
written: spoken:
Our free Single Cell Genomics Day livestream starts tomorrow at 10AM EST! You can ask questions during talks via twitter (include
#singlecellgenomicsday
) and we'll relay them to speakers, see you soon!
So proud of my former student Jonathan Yin!! I know he’s had a blast at
@OctantBio
- thanks for creating such an exciting and nurturing place for him
@srikosuri
So fun to have Jonathan Yin (Yale) join us this summer! After building 100 of thousands of compounds towards our discovery efforts, we've built up a lot of data. He worked on calculating reaction yields for all of our syntheses in more automated ways. [1/2]
And it goes without saying, a HUGE thanks to my mentor and advisor Aviv Regev, who is a bona fide superhuman, but also an incredibly supportive + kind human being!
Fascinating study showing that the microbiome is an endocrine organ. Gut microbes of patients with castration-resistant prostate cancer are enriched for species that produce androgens, and antibiotic therapy delays castration resistance in mice.
Wow. Every cell in our body must be adjusting their state to the thousands of chemical signals in their tissue milieu!
"Sensory periphery separates signals from predictable background via a transcriptional rheostat whose ... state reflects the past and constrains the future".
Olfactory sensory neurons (OSNs) use transcription to make flexible predictions about the environment and dynamically change odor responses, now in our new paper led by the amazing
@TatsuyaTsuka0
and
@davidhbrann
! Thread about what we found 👉1/n
Biden has taken the lead in both Georgia and Pennsylvania this morning as votes continue to be counted:
Results are here as they come in:
Georgia:
Pennsylvania:
Emma Magee is a fantastic RA in the Regev lab who is currently on the PhD admissions market. Admissions committees, get in line 😉! Many thanks to dear friend
@ehshabibi
for his help even during his busiest, Fei Chen who was essential for buffer debugging + all others!
FFPE have damaged RNA. To increase RNA capture, we extracted nuclei *with ribosomes attached* to the nuclear membrane, and confirmed intact nuclei using transmission electron microscopy.
Every time I chop veggies, I’m in awe of their tissue complexity, e.g the bell pepper. Its white sponge-y inner walls — how do they stem from a single progenitor? The outer dimples are surely regulated — they predictably occur 3-4x (via gene expression or mechanics)!
Protein != RNA levels, especially for TFs. Nuclear TFs are impt in fundamental q’s, from cell fates to cell state changes. E.g. cells can respond to environmental signals
by transforming signal into nuclear translocation of TFs. We show how inCITE-seq can be used for that.
Excited to share DIRECTED (Delivery to Intended REcipient Cells Through Envelope Design): a programmable platform for cell-type-specific delivery.
Thank you to
@FrangiehChris
,
@mircoscopy
,
@GuilhemFaure
, Rhiannon Macrae,
@zhangf
1/7
Attempt
#1
for hiring a Research Associate and Postdoc(s) failed in the midst of a global pandemic, so now we're trying again! Short descriptions follow.
RTs appreciated!
Of course, many challenges remain. A key factor is high variability in the prep and storage of FFPE samples. For severe degradation: random primers, polyA'ing short RNA (e.g. SMART-Seq-total), or targeted sequencing (just out) are exciting possibilities.
SO timely! A short teaser preprint from a wonderful collab w/
@dralexswarbrick
lab.
Full protocol available for in supporting material.
snPATHO-seq: unlocking the FFPE archives for single nucleus RNA profiling
@10xGenomics
Excited to see future applications for identifying cell type-specific genes that are associated with TFs directly in tissues, like we show as an example for the mouse brain!
We tested several deparaffinization conditions, and chose xylene at room temp. A side-by-side comparison of RNA detection in frozen vs. FFPE tissue across hemispheres from the same mouse brain showed RNA reduction by ~2.7X
@NTallapragada
@Caroline_Bartma
Depends on which specific fat components, I think. Small chain fatty acids are volatile and could bind ORs, but larger ones (>7 Carbons?) are less volatile so I wouldn’t expect that they trigger ORs in the nose.
Interesting abstract
This was a big team effort with co-lead authors Alexandre Melnikov and Cristin McCabe, our mentors Aviv Regev and Orit Rozenblatt-Rosen, collaborators Sarah Mazzilli
@DrSMazzand
Avrum Spira,
@insitubiology
, and so many others!
@broadinstitute
@BU_Tweets
/end
@Symbionticism
Yes, was obsessed with this early postdoc ! NPs could have evolved from AMPs — early nerve nets evolved in marine organisms that swim in a solution full of bacteria with neurons exposed, perhaps could sense + process external bacterial info.
We tested both plate-based (SMART-Seq2) and droplet-based (10X) snRNA-Seq in the mouse brain. Major cell types are all detected despite reduced RNA complexity.
We also show comparisons to commercially available intracellular buffers as a technical note for the community that capturing intracellular vs. intranuclear targets may require different optimizations 2/n
We applied snFFPE-Seq to a human lung adenocarcinoma sample. Gene detection was sparse, as expected, so we used an atlas to assign cell types to each nucleus. As a sanity check, we reciprocally examined top FFPE-derived marker genes in the atlas.
Some exciting news - I will be joining the Department of Molecular Biology
@Princeton
as an Assistant Professor in January 2021!
The Myhrvold Lab will develop CRISPR-based technologies for detecting and destroying RNA viruses, and for studying RNA more broadly
@PrincetonMolBio
What do we do w/ data? We model gene expression as a linear combination of protein levels to quantify how the expression of each gene changes wrt to protein levels. We did this genome-wide and found that gene modules coincide with specific TF effects!
@meghan_daum
I have a PhD and never ask to be called Dr (only when joking). But expert knowledge is increasingly disrespected in our society and if emphasizing & embracing status of PhDs collectively ⬆️ respect, then by all means! E.g., referring to priests as “reverend” is respectful.
I am infinitely grateful to my advisors
@RoyKishony
, Aviv Regev, and Fei Chen
@insitubiology
.
Fei adopted me during his Year 0 as a junior faculty and has been an unwavering source of support, generosity, and boundless creativity. Without him, I wouldn't be where I am today.
So many directions are brewing, and we can’t wait to start. We are recruiting at all levels. If you are interested, please reach out! We welcome all backgrounds – wet, dry, or amphibious.
Many pathway signatures in databases have redundant and overlapping gene members (!). To avoid scoring the same genes repeatedly, we identified which signatures were redundant by using pairwise Jaccard similarity, and reduced into representative ones.
We work at the interface of tech dev and computation. Our theme is hormone signaling, a fundamental driver of human health in aging, cardiovascular diseases, and cancer. With cutting-edge tools, we're taking on this age-old problem in new ways.
Using the "meta-gene" counts matrix of reduced signatures, we clustered nuclei from lung adenocarcinoma FFPE. This unsupervised approach revealed distinct subpopulations that reflect upregulation of KRAS, downregulation of KRAS, and tumor suppressor PTEN programs!
To highlight the potential for data-driven discovery, we created a computational approach: Gene Aggregation across Pathway Signatures (GAPS). Basically, we make a nuclei-by-signature matrix instead of nuclei-by-gene using known pathway signatures, but after a key step:
Check out our new paper
#GutBrainAxis
#Neuroimmunology
#Microbiota
Many thanks to our collaborators - it's a multidisciplinary project; we wouldn't be able to make it without your help!!
As always, big thanks to my advisors Aviv Regev and Fei Chen, and special thanks to
@naturemethods
for making the review process thorough yet painless!
inCITE-seq also enables protein + RNA profiling in solid tissues. Intact whole cell dissociations are hard for solid tissues. Many original sc tools and CITE-seq were developed for free-floating cells like PBMCs. Single nucleus methods are v. good for solid/frozen tissues.
Finally, a huge thanks to the team! Chris Parkhurst is a postdoc with David Artis at
@WeillCornell
whose other life as an MD/PhD is a pulmonary critical care fellow. He sprinted to ship us samples the day before labs closed + had to treat COVID patients in NYC non-stop in March.
Note: spatial genomics has been making rapid progress for FFPE (e.g. Visium, GeoMx), but they are not truly at single-cell scale *yet*. As these tools mature, we are excited about all the data integration possibilities.
A fav observation was the relationship b/t NeuN protein and its encoding RNA Rbfox3. NeuN is an RNA-binding protein, and these typically negatively regulate their own transcripts. Our unspliced intron counts vs. mRNA for Rbfox3 hinted at this too!
@t_ouspenskaia
they form fused, elongated, tubular networks under certain conditions! i saw a talk on this years ago, and was curious whether it has sth to do with increasing their surface area. here's a bioessay about mito networks:
cc
@LeifLudwig
@laura_boykin
@cambearon
@The__Taybor
Yay glad this resolved :) I’ve known Cameron for years as a colleague and friend, and he’s anything but some “punk” - he’s a very serious and rigorous scientist (also nice person)
Our research spans various areas: spatiotemporal dynamics of ovarian aging (with star team
@jenngarrison
@tammyinlab
@davidsebfischer
+ others), mapping PROTAC response with single-cell proteogenomics, and modeling multi-cell-type state transitions in spatial data.
1/3
I am doing something new (for me) for my talk tomorrow at the
#apsmarch
meeting. I will be using web-based slides, openly available at . If you have any suggestions on the slides, I would love to incorporate them into my talk tomorrow.
#OpenScience
This photo of the sun might not look too impressive... until you realize it was taken at night – not looking up but looking down, through the entire Earth, using neutrinos rather than light. Amazing!
Our work suggests that resistance mutations could be driven to extinction during early stages of infection by designing patient-specific antibiotic cycling strategies, informed by deep genomic surveillance.