RT @luke_phys: Want to work on cutting edge theoretical research to advance our mechanistic understanding of biological processes? Do you e….

RT @ramon_grima: I have an opening for a postdoctoral research associate in my group at @EdinburghUni to develop new mathematical models o….

#PhD opportunity: fully funded PhD position at the University of Edinburgh on 'Modelling stochastic emergence of antibiotic resistance in bacterial populations' with @HelenKAlexander and myself. For more info on the project see here: or get in touch!.

Led by the brilliant @xellbrunet. Excellent collaboration with @group_tomlinson; Tibor Antal (@EdinUniMaths); @ISoriano_Moruno; @NathalieFeeley; @drthorn. Feedback would be most welcome.

New paper! Maximally dysregulating APC should provide the maximal selective advantage to colorectal cancer cells? Doesn't appear so. We quantify 'just-right' APC inactivation, at scale, in a variety of contexts, in our preprint:.

Happy to share that I'll be joining the School of Maths @EdinUniMaths as a Chancellor's fellow (~tenure track assistant prof). Will be continuing to apply probabilistic thinking to genomics. Very grateful for all support and encouragement I've received. Come say hi if in town!.

Highly creative, elegant analysis providing deep insights on DNA damage induced mutagenesis. Please do give it a read. Congrats to @craigandersn, @odom_lab, @S_J_Aitken, @mst_paralogue. It was a privilege to contribute.

Great PNAS commentary piece highlighting our recent work on DNA damage:transcription interactions.

RT @gcaravagna: [ Cancer biomarker discovery is historically mutation-centric, often neglecting the key role of ge….

This work has been a fantastic collaboration with Martin Taylor (@mst_paralogue ), Craig Anderson (@craigandersn ), Sarah Aitken (@S_J_Aitken ) and Duncan Odom (@odom_lab ). We'd welcome any comments or thoughts.

The observed 5' bias in repair efficiency is a consequence of polymerase not restarting, but this pattern wouldn't exist for lower damage burdens. High damage mutation patterns != linear amplification of low damage mutation patterns.

When repair is triggered, RNA polymerases generally don't restart transcription from the repair site; in fact the data is consistent with the polymerases always dissociating with repair 10/n

RNA polymerases frequently bypass damage with triggering repair with lesions transcribed-over 1-3 times before repair. While small alkylation adducts are unlikely to be an efficient barrier to gene expression, they frequently result in mutant transcripts 9/n.

To map the mutation patterns into estimates of how frequently polymerases detect damage and their fate after repair, we developed a stochastic model characterising the interactions of RNA polymerases with damage. Analysing the mouse data through the model we conclude that: 8/n

As expected, higher transcription -> higher repair. More surprisingly, we observed a gradient of repair efficiency moving through the gene body 5' -> 3' for genes with intermediate expression levels 7/n

Lesion segregation patterns offered comparison of mutation burden resulting from template strand damage (affected by TCR) to coding strand damage (unaffected by TCR), allowing quantification of TCR efficiency through genes 6/n.

To put numbers on these outcomes, we analysed mutation patterns from a mouse model of liver cancer induced by an alkylating agent 5/n.

Removing damage = good. Stalling polymerases and stopping transcription = bad. So how frequently is repair initiated?. To minimise disruption of transcription, it would be helpful if polymerases restarted transcribing immediately after repair, how frequently does this happen? 4/n

When RNA polymerases encounter damage, they can stall, triggering transcription-coupled repair (TCR) to remove the damage 3/n.

A huge amount of DNA damage happens to each cell every day (~100,000 DNA lesions). This damage can cause oncogenic mutations and disrupt gene expression 2/n.