
Postdoc @Stanford. Engineering immune cells to fight cancer. Ph.D. @ETH_en | M.D. @CayetanoHeredia 🇵🇪 🇨🇭
Palo Alto
Joined January 2013
- Tweets788
- Following578
- Followers328
- Likes862
Daniel Hoces retweeted
Stellar viewpoint from Ira on the current status of cancer immunotherapy - takeaway message: small biotech and academia will need to continue to keep the candle lit for immuno-oncology, and more mechanistic studies will be needed to find the next big thing nature.com/articles/s41577-0…
Daniel Hoces retweeted
This is pretty exciting, STAT degeneracy producing T cell heterogeneity I think is (was?) one of the major unsolved questions in immunology
Replying to @biorxivpreprint
@biorxivpreprint A Cytokine Receptor Signaling Atlas Reveals How STAT Mosaics Fine-Tune T Cell Function
biorxiv.org/content/10.64898… @Stanford
Daniel Hoces retweeted
There's a subset of women who have four cones (instead of three) in their eyes, enabling them to distinguish between colors that nobody else can.
Most of these women probably have no idea that their vision is unique.
For context: The human retina (usually) has three cone receptors that, roughly, are tuned to sense short, medium, or long wavelengths. The brain "sees" an image by comparing how activated these three sensors are. It is the ratio of activation that determines the color we see.
Hence, if you were to look at a pure orange light, and then look at a light made by mixing red and green lights, the two would look identical to your brain even though they are made in different ways. This partly explains, for example, why our phone screens only need three kinds of pixels (red, green, blue) to convince the brain that it is seeing every possible color.
Now, the genes for the medium- and long-wavelength cones sit right next to each other on the X chromosome. Sometimes, when eggs or sperm are made, they can "swap over" each other and make a hybrid gene that is tuned to detect light *between* those two wavelengths. A man who inherits a hybrid gene on his X chromosome is called an "anomalous trichromat," and will generally have trouble telling apart reds and greens.
But his mother, carrying an ordinary gene on one X chromosome and a hybrid gene on the other X chromosome, is different!
Early in development, cells in the retina randomly silence one of the X chromosomes, and its descendant cells then inherit that selection. Since there are millions of these cells in the retina, each of those cells will express the gene on either this X chromosome or that X chromosome. This means that, in these rare women, their retina will actually contain *four* types of cones rather than three.
In 2010, researchers tested the vision of women who have sons with red-green colorblindness, and soon found a woman -- who they called cDa29 -- that seemed to possess four cone cells.
They tested her child with an anomaloscope, an instrument that projects a small circle split across the middle. One half is lit by a single pure orange wavelength, and the other is lit by a mixture of red and green light. A knob is used to control the ratio between the colors, and the participant is asked to turn this knob until the two halves look exactly identical. The son was red-green colorblind.
Then they showed the mother color pairs that look identical to most of us, but look different to a person with a fourth cone. She could quickly spot color differences that nobody else could. cDa29 was later sequenced, and researchers found that she had normal cones for long-, medium- and short wavelengths, but also a hybrid gene made by merging the first four exons from the middle-wave cone, and the last two exons from the long-wave cone. She can see (and distinguish colors) between these two wavelengths.
Article snippet from a recent New Yorker article where I first heard about this.
Transcription Factors are central to cellular identity and function. The human genome currently contains ~1600 of them. But could there be productive human transcription factors that have not evolved? Out today in Cell, a tremendously talented graduate student Oliver Takacsi-Nagy @Oli_TN, together with the @Satpathology Lab, determined to find out. (1/10)
authors.elsevier.com/sd/arti…
Daniel Hoces retweeted
Incredibly excited to finally share my PhD work with the world! Huge gratitude to the Daniels Lab team, collaborators, and mentors who helped bring this project to life.
Learn more about the work in our tweetorial below 👇
🚨What if we could reliably program macrophage polarization state?🚨
biorxiv.org/content/10.64898…
Macrophages are highly plastic immune cells that perform critical functions by polarizing into distinct cellular states. The polarization state of macrophages can substantially influences the progression of cancers, infections, and autoimmunity. (1/11)
Daniel Hoces retweeted
🔬 We’re hiring again!
Looking for 1 #PhD student to join the #MSCA-DN TRANSCEND network 🧬
Project on unconventional #Tcells in #MultipleSclerosis with research exchanges in
🇩🇰 Eyer’s lab, Aarhus
🇳🇴 @victorgreiff’s lab, Oslo
⏳ Deadline: May 10
👉 euraxess.ec.europa.eu/jobs/4…
The size of new DNA sequences that can be integrated into the human genome is a foundational constraint for engineering and enhancing human cells. In a new collaborative study, we’ve now almost doubled the maximum size of DNA sequences that can be efficiently inserted into primary human cells.
biorxiv.org/content/10.64898…
Daniel Hoces retweeted
LLMs won because they were native to text.
Treating tables as flattened tokens was always a hack.
Structured data needs its own foundation models — ones that understand schemas, relationships, and numerical semantics from the ground up.
That’s where the real enterprise value is.
The next big AI wave won’t be prose — it’ll be rows, columns, and relations.
forbes.com/sites/rociowu/202…
Daniel Hoces retweeted
📢 We are hiring!
2 #Postdoctoral Positions in #NeuroImmunology in my group at @MyUniSR , fully funded for 2 years.
@syis_ecr @y_efis @SanRaffaeleMI @SiicaI
Details below 👇
Daniel Hoces retweeted
1/ There is still debate whether antigen is required to generate memory T cells. I think true memory T cells cannot form in the presence of their cognate antigen.
#immunology
#science
nature.com/articles/s41590-0…
nature.com/articles/s41590-0…
Daniel Hoces retweeted
Happy to share our latest paper on T-cell compartment in children with type 1 diabetes at the diagnosis and one year later. Reanalysis of other available data included. Credit to @vercanie and all co-authors, including great collaborators from @FnMotol.
tinyurl.com/bp99r6ae
Daniel Hoces retweeted
Cool CAR-DCs. #celltherapy science.org/doi/10.1126/scit…
Daniel Hoces retweeted
For readers interested in the potential of in vivo CAR-T therapies in oncology and autoimmune diseases, here's a recent review
rdcu.be/eUJUe
nature.com/articles/s41573-0…
Daniel Hoces retweeted
In-vivo CAR-T is getting hot, fast, and the “how” is now the whole game.
This chart is a clean snapshot of where the field is clustering, and where it’s splitting.
A few observations:
1️⃣Clinical activity is leaning heavily viral, LVV with stable CAR expression, multiple groups converging on T-cell targeting ligands.
2️⃣Non-viral is the wild card, mRNA-LNP for transient CAR, circRNA-LNP, mcDNA, EDV, all aiming for repeat dosing, cleaner CMC, and scalable supply, but the biology has to cooperate.
3️⃣“Stable vs transient” isn’t a footnote, it’s the product strategy, transient may fit autoimmunity and tunable dosing, stable may win where persistence is mandatory.
Bottom line, we’re past “is in-vivo possible”, now it’s “which delivery vector hits the right cells, at the right % editing/transduction, with the least chaos.”
Credit: Cesare Di Nitto on the HR app
Daniel Hoces retweeted
Immune suppression in vivo is antigen specific and obligatorily mediated by antigen-specific Treg cells, not bystander effects. however, many Treg researchers still cling to early in vitro artifacts.
#science
#immunology
cell.com/immunity/abstract/S…
Daniel Hoces retweeted
#ScienceSaturday
❓ How do scientists find the immune cells that are truly fighting a tumor?
➡️ A new @Nature study reveals that powerful tumor-killing T cells may be hiding in plain sight, not as single cells, but in clusters attached to tumor cells or antigen-presenting cells. These clusters, often filtered out during traditional cell-sorting, contain CD8 T cells that are more tumor-reactive, more activated and more effective than typical T cells found alone.
➡️ Using advanced flow cytometry and single-cell sequencing, researchers showed that these clustered T cells carry gene signatures linked to strong anti-tumor activity, high TCR clonality (a marker of targeted immune response), and increased ability to kill cancer cells. In fact, T cells taken from clusters were up to nine times more potent at destroying tumor cells compared to single T cells.
➡️ Even more exciting: when these cluster-derived T cells were expanded and tested in mice, they infiltrated tumors more efficiently and significantly slowed tumor growth, showing promise for improving adoptive T-cell therapies.
🌟 Congratulations to senior author Daniel S. Peeper and the collaborative teams at the Netherlands Cancer Institute, Oncode Institute, VIB Center for Cancer Biology, University of Gothenburg, Harry Perkins Institute and others for uncovering a new path to potentially strengthen TIL therapy and immunotherapy outcomes. @d_peeper @vib_ccb @oncodeinstitute @NKI_nl @uniofgothenburg @PerkinsComms
Read more here: doi.org/10.1038/s41586-025-0…
Daniel Hoces retweeted
Cool. Can we finally let go of the "cancer starts in a single cell" idea?
What’s true in #PancreaticCancer precursor lesions is also true in #ColorectalCancer precursor lesions. @cncurtis @SnyderShot @StanfordMed demonstrate polyclonal origin of adenomas in patients with FAP.
Cancer as a manifestation of convergent evolution!
nature.com/articles/s41586-0…
Daniel Hoces retweeted
Inhibitory PD-1 axis maintains high-avidity stem-like CD8+ T cells @Nature #Germain @NIAIDNews
nature.com/articles/s41586-0…
Daniel Hoces retweeted
Profiling antigen-binding affinity of B cell repertoires in tumors by deep learning predicts immune-checkpoint inhibitor treatment outcomes nature.com/articles/s43018-0…




























