RT @NettaMendelson: New paper from the Tanay lab on longevity is out today in @NatureAging! We employ machine learning to tackle a classica….

RT @unDtpl9Etkp27U8:

The most fun image & video creation tool in the world is here. Try it for free in the Grok App.

RT @ymayshar: Our work on rabbit gastrulation by the Stelzer and Tanay labs is now online! Spicing up the classic evo-devo hourglass theory….

RT @ZMukamel: Our work on the functional role of DNMT3A/3B during early development is finally out!.Grateful for the opportunity to work wi….

TET knockouts cause severe gastrulation defects, but in chimeras, they differentiate (almost) completely normally. Check out how @SaiFeng_Cheng and @mittnenzweig use that to separate cis from trans effects!

Congratulations to my sister and brother in-law! @unDtpl9Etkp27U8 @orielber.And to my daughter who is the lead actor!.

Now on Genome Biology: Code: Docs:

RT @NatureComms: Study investigates DNA methylation landscapes of 1538 breast cancers from the METABRIC cohort and reports epigenomic insta….

Also check out the wonderful blog post by @rnbatra which elaborates more on the emerging model for DNA methylation in cancer: .

RT @NatureMedicine: In our latest News and Views: Marina Sirota and colleagues @UCSF_BCHSI discuss new data on the application of #machinel….

This and more in the paper and the accompanied resources. Now we all need to scale this to single cell resolution….

Such mis-regulation is stochastic and usually non-functional, but the same is true for most genetic (“passenger”) mutations. In some cases however, it can co-opt with genomic events to drive cancer, as suggested by the prognostic value of the epigenetic instability score.

What is then the impact of DNA methylation changes in cancer given the new data and model? Most importantly we think the tumor methylation clock and the epigenetic instability dynamics make cells more prone to mis-regulation.

But after controlling for global effects, we can go back to the genome and robustly detect thousands of promoters and enhancers where the linkage between methylation and expression is specific 𝑖𝑛-𝑐𝑖𝑠 – really exciting!

Now note that these layers represent global effects that affect the entire genome. It is only natural that without context, they were frequently misinterpreted as specific changes at specific genes.

𝐓𝐡𝐢𝐫𝐝/𝐟𝐨𝐮𝐫𝐭𝐡 𝐥𝐚𝐲𝐞𝐫𝐬: DNA methylation changes that correlate with X-inactivation or copy number aberrations. The paper shows how in some cases this is implicated in dosage balancing.

𝐒𝐞𝐜𝐨𝐧𝐝 𝐥𝐚𝐲𝐞𝐫: classical CpG islands that gain methylation. This is highly correlated with a major proliferation gene expression signature. And it is remarkably prognostic. We define it as indicating 𝑒𝑝𝑖𝑔𝑒𝑛𝑒𝑡𝑖𝑐 𝑖𝑛𝑠𝑡𝑎𝑏𝑖𝑙𝑖𝑡𝑦.

This layer behaves like a tumor methylation clock. But it is not the common age clock (. And it is not correlated with mutations, gene expression or prognosis. Just a background process slowly degrading the cells’ epigenome.

𝐅𝐢𝐫𝐬𝐭 𝐥𝐚𝐲𝐞𝐫: most of the genome is losing methylation slowly in a replication dependent fashion. Big blocks that are late replicating lose more methylation (e.g., PMDs . @suscla1 @peterwlaird @benbfly

We infer a model built over layers of epigenetic dynamics. It generalizes years of (sometime contradicting) literature into one common framework: from early reports on estrogen receptor silencing to our days.