I'm excited to unveil a project close to my heart for years now on BioRxiv titled: ‘Clonally heritable gene expression imparts a layer of diversity within cell types’ or ‘How I learned to eschew heterogeneity, and love homogeneity in scRNAseq data’ 🧵1/23
Stop renaming effector T cells just cause you found some of them express some different genes in your 10x data. It’s confusing everyone. They are just effector T cells. Not CD8+GZMB+ cells which also express Interferon genes and KIRS (CD8GINFKIR)
Paper is out today! Took 2 years. Cell Systems was great review process all around. Spent some time at Science but editor quit mid-review and one reviewer didn't see the point 'we know cells have memory, is it just that this shows more of it?!'
T cells have no Hayflick Limit and most importantly: 'Our results do not reject the concept of T cell stemness, but dissociate self-renewing potential from markers previously associated with the ability to maintain longevity and proliferative fitness.'
Requirement that 3 independent reviewers all have positive feelings about a study that challenges existing dogma (of which any of them may have built their careers) all but guarantees that no truly novel research will be published in competitive journals.
I can’t wait till scRNAseq or spatial goes clinical. I can just imagine…
Doctor: “I don’t know how to tell you this…”
Patient: “just give it to me straight Doc”
Doctor: “we found…. Heterogeneity “
I read lots of old papers and I’ve started to really feel that “less is more”. Looking at old vs new slide decks I feel the overload of information in modern pres/papers is not translating to a proportional increase in insights. Am I alone?
Every American that’s ever moved to Europe has at one time had the thought “if I could just find a way to bring burritos to these people I could be a millionaire “
Time limits are stupid. Full stop. We should be promoting people with management skills and maturity to teach/mentor. Not people who can crank out a few “high impact” papers in 5-7 years. And one wonders why there’s issues with cheating and poor management/mentoring…. 🙄
Congratulations to all
@ERC
Start grantees. Also a positive thought to the majority of applicants that were not awarded or could not apply because of the (short) time limitations after PhD! Personal take on this 👇
Stop making me write a one sentence summary of my paper - I already did that, it is called the 'Title'. It's embarrassing to have to rewrite the title with different words and pretend like it is a new thing.
Now you can get a Nature paper for doing 1-3 million scRNAseq cells… so if you’re planning to do as much scRNAseq as possible for your project I’d assume you get a paper in for 10-20m cells in about a year
I have been thinking more about publishing, peer-review, and the state of academia (from a middling career scientist) and I have a possible solution I would like to explore in the coming year - follow along if you are interested, I hope to make it worthwhile! 🧵1/16
As I prepare to teach T cell immunology this morning - I would like to reflect on the fact that one of the most influential papers in the field of T cell biology was published in the Journal of Immunology - not CNS journals.
I predict the next 10 years will see a massive overhaul of the way we understand how cancers develop and progress. For me the biggest finding of the past 5-10 years has been the finding that typical cancer mutations are likely common in healthy people
If a highly reproducible result is published on
@biorxivpreprint
but languishes in peer review for a year or more. Does it 'matter' if it is peer reviewed or not? If others can reproduce it is that not a form of peer-review? Then who cares what corporate journal owns it?
I have been immersed in the concept of ‘non-genetic individuality’ for the past few months while writing/revising papers and wanted to do a thread summarizing the field as I think it may be interesting for some! So here goes (it's LONG!): (1/27)
Immunologists - what in particular do you think you will gain by using scRNAseq that you wouldn’t with a well designed set of FACS panels? Provide examples. Is it worth it?
Any rule that limits the amount of time a PhD student can pursue a PhD to less than 5 years is cruel in most fields of cell/mol biology. It takes me around 1 year just to go through the publication process for 1 paper. Rules like this are incompatible with any work/life balance.
I had no idea that my 'grinding my gears' on T cell types would resonate so much... Stick around next week and I will entertain you with: 'Those aren't new Dendritic Cells' or 'I don't think T cell clones can have 2 beta chains' or perhaps 'NK cells: Do they really exist?'
People still are doing velocity on unrelated (not clonal defined/barcoded) cells to imply relationships despite many days of separation in putative past division times in vivo... and publishing it in Cell as main figures. In this case pooled from different mice I believe.
The worst thing about the single cell methods field is how often the work purports to change the way we understand biology while failing to cite decades of careful, less flashy work which these novel methods summarize and present as new. I may be guilty of it too…
One of the easiest and best things you can do as a scientist is to spread other peoples interesting studies around (on Twitter for example) - this is especially true for papers that aren’t in HI journals
PhD students in biomedical research should be required to take a semester long course on human disease pathology. It’s so valuable to see what the human diseases we model look like in reality - especially commonalities and *gasp* heterogeneity (not the cool single cell kind)
Why do cancers express so many Collagen genes - I want a molecular answer - not 'because they need to remodel ECM'. There must be a common pathway upstream of these genes that gets triggered in tumours right?
As promised here is a link to GitHub which should be a public page with my short lecture (39 slides) and 'outline/cheatsheet' - there is a ppt and pdf of the lecture
I teach basic T cell immunology every Fall to 2nd year undergrads at KI with no Immunology background (other than being confused for a few weeks by
@Agingfencer
). I will post a link to a place where people can have my slides and lecture outlines soon.
The solution for everyone is staring us in the face: Step1 - submit to your favorite HI journal, Step2 - go through review, Step 3 post editorial acceptance + all reviews to bioRxiv. Step4- decline to publish. Voila - no fees
I’ve published over 30 papers in what people consider “good” journals and on a weekly basis I wonder if the ideas/experiments I’m running at any given time are totally foolish. Never give up 😅. Doubt is totally normal
These are *enormous* numbers, making IJERPH one of the largest open-access mega-journals in the world.
So now you know why this is BIG news.
The second largest journal in the world just lost its impact factor.
I teach basic T cell immunology every Fall to 2nd year undergrads at KI with no Immunology background (other than being confused for a few weeks by
@Agingfencer
). I will post a link to a place where people can have my slides and lecture outlines soon.
@edyong209
@RheaBoydMD
Most people I have encountered in Sweden who are hesitant to get vaccinated seem to have been scared by the message 'it can't be safe, it was too quickly developed.' It is critical to point out to all that it is built on decades of knowledge and a relative reduction in red tape
I want to share a story of something that really impacted my career when I was a 20 year old undergrad - and I think it could be useful for young students and for powerful academics alike! 🧵
I met with a company today that will essentially produce a cell atlas for you from frozen tissue pieces you send for a fee… including ATAC, RNA(nuclei), spatial, and all analysis. You just send frozen tissue. I kind of liked that 😊. 4-5 samples with 4-6 week turnaround.
I think it’s unbelievably sad that my field’s flagship journal (JI) has the same IF as the predatory journals that offer me accelerated publication of my preprints despite these journals not being on topic. IF is a worthless metric
#Goodhart
’s law
After wasting an hour with google I am now convinced there is no such thing as 'cancer stem cell genes'. People just take the genes they find and run them against any stem cell lists and call the overlap CSC genes. It's the Wild West in Omics - a signature is worth nothing
I just discovered this paper:
David Baltimore theorised how TdT would create TCR diversity years before somatic recombination was demonstrated by Tonegawa..
I want a collection of these sorts of theory papers to give to my students in the future!
This paper is really interesting and gets at the heart of what it is to be a mature effector T cell vs an 'immature' stem cell memory T cell... I must have overlooked it but it raises some interesting philosophical points about T cell maturity.
If you want to rejuvenate the immune system you better start by fixing the thymus. If you want to fix the thymus you better start by understanding why it disappears. If you want study that you better look at long-lived mammals where it disappears in the first quarter of life.
I put some T cells in a well in ‘82 - threw on some cytokines just to see what they would do. I watched em differentiate and called them Th2. Oh a oh - it was a big hit, oh a oh - reviewers loved it
I predict that in 2022 we will see a lot of immunology papers finding 'continuums' and 'spectrums' of cell identities with their new favourite Omics tools. And in 2023 we will see heterogeneous spatial organisations of these spectral populations.
Stoner thought of the day: “what if the development of multicellular organisms required the mechanisms that give rise to tumors and our ancestors were malignant eukaryotic colonies…”
This new paper on paternal transgenerational inheritance of immunological phenotypes in mice may seem like magic to immunologists, but it builds on decades of work in other systems and organisms showing similar things 🧵 1/21
@PhilipAhern
@gerdosi
One thing my friends who debate vaccines vs natural infection miss is the fact that natural infection likely generates a much more comprehensive and diverse repertoire of T and B cells than the RNA vaccines against spike
A classic, but underappreciated paper characterizing the presence of activated T cells and Tregs (pre-knowledge that this is what they are) in the human fetus by arguably the greatest living immunologist Max Cooper:
The first in a series of voluntary reviews of
@biorxivpreprint
. The paper is titled 'Plasticity and lineage commitment of individual Th1 cells are determined by stable T-bet expression quantities' from Ahmed Hegazy and Max Loehning at
@DRFZ_Berlin
I think an excellent first year PhD course in biomedicine - 1st semester- would be to teach a history of medical research and society with required reading: “microbe hunters”, “the billion dollar molecule”, “arrowsmith”, “bad blood”, and “the emperor of all maladies”
I find it humorous that the “publish or perish” model that was made an official metric of merit by the invention of the H-index will eventually destroy peer review under the weight of the number of papers people try to publish to advance their careers
why is it breaking news that there are memory T and B cells to an acute viral infection up to a year after the infection? What is the source that says that a viral immunity can disappear after a year?!
Next year I am going to apply for an ERC grant just so I can come up with a ridiculous acronym and force people to talk about it. 'Spatial Transcriptomic Understanding of Pervasive Infectious Disease sequencing'
Can we get a solid definition for what a “memory T cell” is?
Is it defined by marker genes/proteins?
Recall response (if so is it black/white or shades of magnitude)?
Longevity post initial immune response?
Persistence through multiple responses?
TP53:
- Discovered 45 years ago, most cited gene all time
- No therapies
- 500 million people currently living will die of TP53 mutant cancers without new therapies
Our Preprint:
- A general strategy for TP53 missense mutant cancers (majority) with prototype small molecules
“Spatial mapping using four different technologies revealed an unexpectedly rich ecosystem of tissue-resident immune cells”. Just once I wish an atlas would find heterogeneity and not be surprised by it 😄
Some labs really excel at publishing high impact papers with a ton of experiments in the p=0.01 to p=0.05 range... I am not a stats expert but how many t-tests would one expect to be in that range for a study with 30-50 statistical comparisons in their main figures...
I don’t study fetal immunology anymore - but I used to at a time when nobody studied it. Sometimes I wish I’d continued (was told I’m stupid to stop) but now that it’s fashionable I’d like to share our most important finding - fetal and adult immune cells are different.
The standards for scientists to generate illustrations for the top journals are getting a bit out of control... (not commenting on quality of research - purely on the astonishing beauty of the illustrations herein)
I find studies that look at effects on short-term dietary interventions on immunity really fascinating - a big one just came out and I guess it will get a lot of attention (from a very good lab). I wonder what people think about how strong it is?
Did the Tabulis folks find the first Macrophages, mast cells, NK cells and ILCs with rearranged TCRs? If so I take back everything I have said in the past. This is truly novel biology. Am I wrong? Please correct me if so...
2024 was an year of hope - armed only with 500,000USD of 10x kits and a penchant for the word 'spatiotemporal' we set off to map and chart the unknown at ever increasing resolution.
Maybe the reproducibility crisis isn’t due to a need for a 50 page addendum with all data, reagents, and weather conditions on the day of each experiment… maybe it’s because the effect sizes aren’t all that great to begin with
And the people bowed and prayed... to the UMAPs that they made... And R flashed out its warning - to the trajectories that were forming.. and the genes of the prophets were inscribed on the tree inference, no reference... echo - the sound - of atlas.
@OdedRechavi
Well, it probably would have been a UMAP, but he almost certainly would have invented a new way to process single beak parameters and called it FINCH-beq.
These guys got MALAT1 in their scRNA DE analysis and just went with it 😃
'Chromatin-associated lncRNA Malat1 regulates Th17 effector program and promotes intestinal inflammation'
Does anyone know good examples of cancers/or “pre-cancerous” cells which actually improve the function of the tissue they reside in in a beneficial way prior to collapsing it in a malignant way?
I love how we are expected to upload terabytes of data and code and descriptions but journals are still like - 'guy, you are going to have cut 500 words and 3 references out that explain that data, we simply cannot spare the space.'
Wait - have people only been sequencing the exome of cancer patients? I mean this sounds interesting but how has nobody ever looked at enhancer mutations before?
Other than virtually guaranteeing funding at many places the main reason to publish in CNS is that otherwise it’s incredibly difficult to get visibility. That’s why I use this site. It’s the only way to get eyes on your work on a global scale without relying on CNS
Paper officially out now in Cell Reports - big shout out to
@jakmic2003
@joanna_hard
@laurent_modolo
and
@FrisenJonas
et al.
Divergent clonal differentiation trajectories establish CD8+ memory T cell heterogeneity during acute viral infections in humans
Does anyone else watch nature documentaries and think 'man... I want to run RNAseq on that weird toads gonads to figure out how they can regulate reproduction based on environmental fluctuations...' Is it just me?
Working in a large neuroscience group I’m always impressed with the overlap of key TFs between neural stem/progenitors and T/B cells 😅. (Eomes, ezh2, sox4, bcl11a/b, gata3.. likely more!)
I publish in
@biorxivpreprint
because I love writing up my work. I discuss my results openly because I enjoy productive feedback. I submit to glam journals because an old guy told me he needs a way to measure my worth and doesn’t have time to read my work
@cshperspectives
What would happen if you made a CAR-NK cell with a TCR specific for a given antigen? CD8 T cells can express KIRs and other inhibitory receptors and presumably still use TCRs - would an NK engineered to express a TCR just be a *gasp* CD8+ T cell?!
The best thing about studying cancer gene expression is that literally every gene you google comes up as 'this gene is critical for cancer progression' AND 'this gene is protective against cancer progression'. So you just have to pick your GO strategy and can say anything ;)
I’ve never been a fan of hiding data till papers are accepted. If I have an idea and think someone else is better situated to do it, it’s enough to see it done well. I hope that someday there’s an H-index for “good advice/inspiration”. Not all ideas are “cheap”
😉