Check out our new study in @ScienceMagazine, where we take on a 100-year-old debate: what’s the role of aneuploidy in cancer?

We discovered that genetically removing extra chromosomes blocks cancer growth - a phenomenon we call “aneuploidy addiction”. science.org/doi/10.1126/sc… In the 19th century, pathologists observing cancer cells under a microscope noticed that they frequently underwent weird mitoses. The chromosome bodies visible in these cells were not equally divided between daughter nuclei - in other words, they were aneuploid.

Very excited to share a new paper from my lab: using a set chromosome-engineering tools, we show that cancers are “addicted” to aneuploidy. If you genetically eliminate single aneuploid chromosomes, cancer cells totally lose their malignant potential!
biorxiv.org/content/10.110… To back up, for many years researchers have used the standard tools of molecular genetics to learn about the function of individual oncogenes and tumor suppressors. We can easily over-express, mutate, or knockout genes like KRAS and TP53 to study their biology.

If you choose to transfer a manuscript between Nature-family journals, you can consult a web page that lists the acceptance rates for 124 journals published by the Springer Nature Group.

I haven’t seen this data circulated before, so I copied it to share here: According to this data, "Nature" is not actually the most selective journal. Nature Med, Cancer, and Human Behavior all have lower acceptance rates.

Interesting new paper with an important message - normal clone-to-clone variability is an important confounder in CRISPR experiments.

nature.com/articles/s4159… Westermann and colleagues were studying a gene believed to regulate YAP1 expression. They made two CRISPR knockout clones in the gene. Unexpectedly, they found that one KO clone upregulated YAP1 while one downregulated YAP1!

New from my lab: we show that a clinical-stage oncology drug from Eli Lilly is mischaracterized, and its true anti-cancer target is EGFR. We also show how in vitro drug assays can be misleading - cellular+genetic methods are needed to determine drug MOAs.
biorxiv.org/content/10.110… Ralimetinib (LY2228820) was developed by Eli Lilly as a p38a inhibitor. Thus far, it has performed poorly in clinical trials, with basically no tumor regressions attributed to its effects:

Out now from @joans and me - (nearly) everything you know about survival analysis in cancer is wrong.

cell.com/cell-reports/f… We conducted a comprehensive analysis of genetic, epigenetic, and transcriptional features linked with patient outcomes across 10,000 patients and 32 cancer types.

A web portal for exploring this resource is here: tcga-survival.com

New paper from @joans and me! A pan-cancer, cross-platform analysis identifies >100,000 genomic biomarkers for cancer outcomes. Plus, a website to explore the data (survival.cshl.edu) and a (controversial?) discussion of “cause” vs. “correlation” in cancer genome analysis.

https://twitter.com/biorxivpreprint/status/1399950911573737474

We used every type of data collected by TCGA (RNASeq, CNAs, methylation, mutation, protein expression, and miRNASeq) to generate survival models for each individual gene across 10,884 cancer patients. In total, we produced more than 3,000,000 Cox models for 33 cancer types.

In a blinded name-swap experiment, black female high school students were significantly less likely to be recommended for AP Calculus compared to other students with identical academic credentials. Important new paper from @DaniaFrancis:

smith.edu/sites/default/… Some background: one of the best ways to collect real-world evidence of discrimination is through name-swapping "audit" studies. In these experiments, people are presented with job applications, resumes, mortgage applications, etc., that are identical except for the name…

Angelika Amon passed away this morning. She was the greatest scientist I’ve ever met. This is a huge loss for her family, her friends, and for every biologist. As a grad student with Kim Nasmyth and then an independent fellow at the Whitehead, Angelika changed our understanding of the cell cycle.

In two weeks, the Nobel Committee at the Karolinska Institute will award the 2020 Nobel Prize in Medicine/Physiology.

Who will win? We don’t know for sure - but I think that we can make some educated guesses. Science is dominated by a phenomenon called “the Matthew effect”. In short, the rich get richer. Getting one grant makes it more likely you’ll get the next. Winning one prize makes it more likely you’ll win another.

en.wikipedia.org/wiki/Matthew_e…

Here are the award rates for 11 different postdoc fellowships in 2019.

There’s a huge variation in success rates: four different organizations fund fewer than 6% of applications that they receive, while the success rates for the K99 and F32 are >24%. To back up - my appointment at CSHL let me run a lab without doing a postdoc, so I never had the experience of applying for these grants. To help out my current postdocs, I wanted to make up for my lack of experience by doing some research.

Question: can anyone name a paper whose findings were challenged by a “matters arising” or “technical comment”-type rebuttal, but subsequent research proved that the original paper was actually correct? One example: Charles Sawyers published that leukemia patients who relapsed on Gleevec developed ABL-T315I mutations.

Science then published 2 technical comments reporting that other groups didn't find this mutation in independent patient populations:

science.sciencemag.org/content/293/55…

What happens to a paper submitted to a top journal?

Among a set of manuscripts sent out for review by Cell in 2018:

-33% were published in Cell
-26% were published in another Cell-family journal
-7% are still under review at Cell
-The median time to publication was 391 days To back up: in 2018, Cell started the “Sneak Peek” program, in which authors had the option of posting a preprint of their manuscript if it was sent out for review by a Cell-family journal. cell.com/sneakpeek

Two whole-genome CRISPR screens for SARS-CoV-2 resistance are on bioRxiv.

**Among the top 100 hits in each screen, 99 are non-overlapping.**

Could cell type-specific differences could explain this discrepancy? And if so, what’s the “right” way to study SARS-CoV-2 in culture? A few thoughts: Wei recovered ACE2 as their #1 hit, which is strong evidence in favor of the biological validity of their screen.

(Heaton didn't recover ACE2, which they suggest is because their cells transgenically express ACE2 cDNA, though I don't get why that should matter.)

Here are the publication records and research topic areas of 63 faculty candidates in the life sciences who interviewed at R1 institutions in 2019-2020.

70% have a first-author paper in Cell, Nature, or Science, 22% have a K99, and 30% have unpublished work on bioRxiv. Methods: faculty candidate seminars were found on public departmental websites. Pub records were acquired from Google Scholar or Pubmed, and K99's were found on NIH Reporter. I did my best to summarize research topics by reading the abstracts of a candidate's recent papers.

Our new preprint on #COVID19 with Stefan Pöhlmann, Markus Hoffman, and @joans is up. We show that proteases other than TMPRSS2 are capable of promoting SARS-CoV-2 uptake, but camostat (and its active metabolite) can still inhibit their activity.

biorxiv.org/content/10.110… TMPRSS2 is commonly described as “necessary” for SARS-CoV-2 activation. Many papers look at the expression patterns of ACE2+/TMPRSS2+ double-positive cells, ostensibly to shed light on which cell types are vulnerable to coronavirus infections. But it isn’t that simple!

In response to my public bet challenging Michael Levitt’s dangerous COVID19 tweets, someone I’ve never spoken to DM’d me the following.

I want to explain why this person is wrong, and it’s important to call out misinformation, even at the risk of “looking gross”.

https://twitter.com/JSheltzer/status/1287084519846293505

One of the giants in the field of cancer genetics is Peter Duesberg, a member of the National Academy and a professor at UC Berkeley. He did pioneering work on SRC and retroviral oncogenes in the 70’s. Now he works on aneuploidy and cancer. BUT…

en.wikipedia.org/wiki/Peter_Due…

Rosalind Franklin's 100th birthday is this weekend. I've seen many people claim that she couldn't share the 1962 Double Helix Nobel Prize because she died in 1958.

This is false. Dag Hammarskjold won the prize posthumously in 1961. 'No posthumous prizes' wasn't a rule until 1974 Several people responded to this tweet by claiming you couldn't be nominated posthumously. This is also incorrect. Many examples - pathologist Katsusaburo Yamagiwa was nominated 6 years posthumously.

@neurograce @IndyMicroGuy @SteveMSweet @matthewcobb

nobelprize.org/nomination/arc…

My lab’s new protocol on CRISPR-based competition assays to identify cancer-essential genes has been published in Bio-protocol: en.bio-protocol.org/e3682. It’s a straightforward and flexible approach - I regularly teach high schoolers how to do it. At the same time, it has exceptional precision and recall - very important given all of the recent issues in cancer reproducibility - nature.com/articles/s4157…

It's important to note - our review focuses on the *technical* aspects of cancer drug target validation. But, there are also very important socio-cultural aspects of cancer research that are worth keeping in mind:

https://twitter.com/JSheltzer/status/1273993873145106432

People patent cancer-related discoveries and spin out companies based off of what they find in lab. Scientists can earn life-changing amounts of money from this (and, because of the current start-up ecosystem, it doesn’t even have to work for you to cash in!).

New from my lab in @NatureRevGenet - we take a deep dive into the tools and platforms that can be used to discover and characterize cancer drug targets. nature.com/articles/s4157… Work from my lab and from several others has demonstrated that many genes previously reported to be cancer-essential are in fact dispensable for cancer growth.