Sunday, August 27, 2017

A return to blogging and two new papers: Experimental measurement of evolutionary games and Evolutionary instability in collateral sensitivity networks

Well, the last year has been hectic. I moved my clinical practice to Cleveland Clinic, wrote and defended my thesis (corrections pending...) and have started to grow a research group here in the department of Translational Hematology and Oncology Research. I will begin asking each of the new group members to introduce themselves with short posts here soon, and hope to have at least bi-weekly update posts starting next month.

Before then however, I'm excited to highlight that the two posters that I highlighted on this blog just as I moved, are now full manuscripts. The first, led by +Andrew Dhawan studies how drug sensitivities change over the course of treatment, and even during drug holidays.

This work, which appeared in Scientific Reports, has gotten some attention and we were asked to write a more clinical follow on for Oncology Times called "Evading therapeutic resistance through collateral sensitivities: a paradigm shift?", which you can read here.

The most exciting result from this work was the idea that we need to think about collateral sensitivities a bit harder before we translate them directly to the clinic as they are dynamic even on very short timescales. The full paper, "Collateral sensitivity networks reveal evolutionary instability and novel treatment strategies in ALK mutated non-small cell lung cancer" can be read here:

A tantalizing piece of info here too was the not only did drug sensitivity change over the course of treatment, but so did radiation sensitivity...  More on this later.

The second project, an experimental method to directly measure the evolutionary games cancer cells play during the evolution of resistance has just yielded a new pre-print from the group, led by +Artem Kaznatcheev . Readers of this blog and #mathonco work in general will know that we've been working on evolutionary game theory and cancer for some time - really work started by +David Basanta. David and +Alexander Anderson and +Artem Kaznatcheev and I have now published something like 10 total papers between us on cancer and game theory ranging from studying how hormone therapy timing should work in prostate cancer to how we should think about how our drug scheduling affects tumor composition  and even more abstract ideas like how local cell topology affects evolutionary stable states and dynamics.

Transforming the payoff matrix using the Ohtsuki-Nowak transform allows an understanding of how spatial organization (locally) might change the game... (See Intercace paper linked above)

At issue is that the payoff matrix, the heart of evolutionary games, is usually invented rather than parameterized in any meaningful way. And even when it is it is done indirectly (from literature, or disparate measurements...).  To address this, +Artem Kaznatcheev came up with a clever experimental method to directly measure these games, and we found that the qualitative nature of the game itself can be changed!

So, if this piques your interest, wander over to the bioRxiv and check out the pre-print. With any luck it will be appearing soon in the pages of your favorite journal.

Here you can find "Cancer associated fibroblasts and alectinib switch the evolutionary games that non-small cell lung cancer plays"

OK, see you soon!  Happy reading.

Wednesday, January 25, 2017

Society for experimental biology meeting, Oxford, September 2016

Having just handed in my thesis, after many travails, I am eager to get back to blogging regularly, and to keeping a record of my thoughts.

To begin, I'm going to start by addressing some back-log. So, here, I'd like to briefly describe a great meeting I was able to attend this past September in Oxford. Specifically, 12-15 September, at Lady Margaret Hall in Jericho, I attended a meeting hosted by the Society of Experimental Biology, and organized by my two friends, Ruth Baker and Alex Fletcher - both of whom I know from my own time in Oxford.

For nitty gritty details of the meeting, you can still see the speaker list and abstracts that were presented here:

I'd like to take a moment, instead of the nitty gritty, to describe what were some really nice points (outside the science itself - which was also great) about the meeting that I'd like to carry forward to any meeting I'm lucky enough to organize.

First, the size and scope of the meeting. I often find when I go to meetings like ASCO or ASTRO or even the larger mathematical biology meetings like ECMTB+SMB joint meetings, that I'm overwhelmed by parallel sessions and too many people to possibly wrap my head around.  There is often so much going on that I almost would rather isolate myself and speak only to my closest collaborators - which sort of defeats the purpose of the meeting itself. This meeting, called 'The Tisuee Issue', struck a very nice balance on this point. The size of the full attendance, which I would estimate around 50, was just right. It was large enough that there were people I didn't know, and single voices couldn't take over conversations, but small enough that it wasn't overwhelming, and I felt I could branch outside of my normal circle of friends.

I must admit, that there were also a number of my friends/collaborators gave me some confidence to branch out somewhat. Further, the organizers did a clever thing, which was to break us into discussion groups to discuss aspects of multi-scale (and multi-disciplinary) modelling ranging from education of multi-disciplinary scientists to mathematical limitations of these models. The small group discussions were then brought to the larger forum for a full group session. These sessions were great to break the ice between folks that didn't know eachother, and also allowed more junior members opportunities to present to the larger group.

I think the nicest thing about the meeting, however, was that there were two groups there who don't usually come together.  Namely, developmental biologists and oncologists. There are many commonalities between the disciplines, but there is little cross-over. This meeting allowed us to get to know one another, to see how we were using similar techniques for different problems and to learn some new techniques as well.  While I didn't understand all of what the DevBio crowd was talking about, I was able to appreciate the methods and see new ways to apply them to my own work.

On the whole it was a great experience, and we owe Alex and Ruth, and the Society for Experimental Biology a lot for a great meeting.  Hoping to have a repeat in a couple years time!

We also had some great tweeting going on, which you can see in this storify:

Sunday, July 10, 2016

A big move and a new research group. Join us!

My decision to leave the IMO and Moffitt Radiation Oncology was one of the hardest professional (and personal) choices I've ever made. I've made friends I won't ever be able to replace, and learned lessons I'll never forget. However, for many reasons, including family and professional and personal growth, it is time for me to move on. I'm very lucky to have found a post in my home town at the Cleveland Clinic doing exactly what I want to do. They have hired me be a physician-scientist, to start a research group in the department of Translational Hematology and Oncology Research (yes, they call it THOR - how awesome) studying the evolution of resistance, and to continue my work as a sarcoma radiation oncologist.

I'm lucky to be joined by two brave souls who will begin their DPhils under my co-supervision between Oxford and my lab at the Cleveland Clinic. +Artem Kaznatcheev , a theoretical computer scientists and mathematician (who is also a prolific academic blogger - see his blog TheEgg here) will begin in the department of Computer Science this Michelmas with Peter Jeavons, David Basanta and myself supervising. Artem's thesis project will focus on the evolutionary process, but beyond that we're not sure - we'll find something that piques all of our interests I'm sure. Artem's work with David and I to date has included attempts to bring spatial structure into evolutionary game theory using the Ohtsuki-Nowak transformation. We found, broadly, that game dynamics can change significantly based on local neighborhood size. This finding has, I think, broad implications for understanding spatially heterogeneous tumors using game theoretic methods.

and more recently using time lapse microscopy to quantify competition dynamics in vitro - something I'm presenting a poster on at ECMTB2016 later this week.

There are more ideas than time, in general, when working with Artem, so I'm quite keen to begin this new journey and strengthen the connection between computer science that we began with +Dan Nichol doing his DPhil with Pete Jeavons and +Alexander Anderson - which we kicked off with Dan's evolutionary steering paper.

I'm also very luck to have +Andrew Dhawan joining, who will begin his DPhil in Oncology with Francesca Buffa and Adrian Harris. Andrew's DPhil will be slightly more prescribed, it being under a CRUK studentship. He will be focussing on building a comprehensive map of the transcriptional response to hypoxia in (breast) cancer. Many of you will note that this is reasonably far from my expertise, but I think it will be a great opportunity for me to hone my Next Gen Sequencing chops, and also to inject some theoretical thinking into Oxford Oncology.

Andrew has spent this summer, since he finished medical school in Ontario, Canada at Queen's University School of Medicine, working with me and +Andriy Marusyk on collateral sensitivity in non-small cell lung cancer. He is attending a summer school (sponsore by the Cancer Research UK coincidentally) on ecology and evolution in cancer - annoying scheduled the same week as ECMTB. He'll be presenting the following poster there:

On another random note: it turns out, unbeknownst to any of us, Andrew and Artem did their undergrad at the same institution, in the same department! They both went to the 'MIT of Canada' in Waterloo. Small world.

Anyways, I am immensely looking forward to this move - the next few weeks will see me at ECMTB, then home to Tampa for one day, and then the move Cleveland, where I start August 1st. I am also looking for someone interested in joining the team as a post-doc who wants to work with us to make a difference in patient's lives through the study of evolution in cancer (or bacteria) with primarily mathematical methods.

If you're coming to ECMTB, look me up!

Friday, July 1, 2016


Last night +Alexander Anderson hosted a going away party for me. It was a lot sadder than I thought. I am going to dearly miss the people in the IMO, and Moffitt Cancer Center in Radiation Oncology and elsewhere that I've worked with over the past 7 years. While there was a lot of silly-ness, in particular some of the funny shots of me and others from IMO over the years displayed on the big-screen:

There were also some sad farewells

The whole evening reminded me how strongly I feel about the work we're doing in the IMO, and also of the courage it took to create such a place. Originally started by +Alexander Anderson and Bob Gatenby the IMO, when I joined was about 5 people: the two fearless leaders and +David Basanta , +Edward Flach  and +Kasia Rejniak. In the intervening (only 6) years, it has grown to nearly 25 people, with 5 faculty members. I say it took courage to start, because at the time it began, mathematical oncology was nearly an unknown phenomenon. Since, however, it has become much more accepted - supported by specific initiatives from the NCI like the Physical Sciences in Oncology Network (formerly PSOC, now PSON) as well as the Integrative Cancer Biology Program.

The work is meaningful to me on many levels. First, the fact that we're chasing down the fundamental principles of a disease which has proven largely impenetrable for most of human history is intellectually satisfying. But moreso, for me at least, it brings hope to my time in clinic, when times become difficult. The fact that I can take that energy back to the lab lifts me back up and drives me to work harder, and also let's me reassure my patients that we're doing everything we can, both for them and for the future.

Moffitt started a new campaign, called the community of courage, to allow researchers and clinicians to talk about what courage means to them. I was flattered to be asked to join this campaign, and through it talked both about the courage I see in +Alexander Anderson and his group to go outside the norm in mathematical onoclogy, but also the courage I see in my patients, who choose to get up and LIVE every day, even in the face of difficult odds.  They made a little video, which you can see below, and also wrote a nice article, which captures what I've said here quite a bit better than I have.

You can find the article here:

Next post there will be a post-doc advertisement attached...  so if you know someone who is interested in a post-doc position in mathematical oncology, or in studying the evolution of resistance to antibiotics, let them know there's a new lab coming to the +Cleveland Clinic - mine!  Oh, and also that Cleveland Rocks...

Thursday, June 16, 2016

The end of one road, and the beginning of another.

One month ago I sat for my oral boards for final certification by the American Board of Radiology. This exam represents the final hurdle in what is (normally) a decade long road after college. It should be 4 years of medical school, then 5 years of residency and one final year of independent practice. During this time, you take innumerable exams: the USMLE Steps 1, 2CK, 2CS and 3 followed by the ABR radiation physics and biology exam, then written clinical exams, followed at last by the oral boards.

I found it interesting that during the time leading up to residency (during undergrad), we have been selected based on who performed the best. This includes striving for As in college (and high school...), the highest #MCAT score you can manage, and then doing as well as you can on 'the Steps' so as to assure the residency you desire. Once you make it to your speciality training however, this turns on its head. The desire to perform your best is replaced by the abject fear of failure. This fear basically ramps up higher and higher until by the week before the oral exams, everyone in your discipline is just standing at the edge of the abyss of self-doubt (at least I was).

The change is from filter to pump. At the beginning, the desire is to filter folks out, but once the desired level of rarification is reached, they don't want you to fail any more - they want to pump you through. It was a long road, and one which I am extremely happy to be done with.

I will resume this stream next week with an update on my research and what I think is a nice result about stem cells and phylogenetic trees...

A final thought. I was speaking with a friend and mentor of mine recently about the transition from resident to faculty. He said when you finish residency you feel like you've finally climbed to the top of the mountain. But, in academic medicine, you realize that the mountaintop that you just summited is actually the bottom of an even bigger one. Now you are no longer competing against your peers and co-trainees, but instead you are competing against those who trained you. In the ridiculous funding climate, this is a hard truth.  So, while I feel good to have gotten to where I am, the climb is just beginning.

Wednesday, January 6, 2016

Rotating student: Jeff Peacock - working on evolution of resistance in ALKmut NSCLC using CRISPR

+Andriy Marusyk and I have a medical student rotating through our collaborative lab for the next couple of months, here's an intro with a couple figures from a recent grant of mine.

Also, here's Jeff:

As an aspiring radiation oncologist, what originally drew me to the field is its investment in scientific research. My name is Jeffrey Peacock, and I am a visiting 4th year student from UCF who is working in Jake’s lab for the first 2 months of 2016. I am excited to work on some really cool projects that Jake has started. I have spent a few years in a wetlab, both during undergrad and before and during medical school, performing genetic engineering on bacteria and yeast to produce commodity chemicals. One experiment that I designed was a directed evolution study to increase turnover of a bottlenecked enzyme in a metabolic pathway. I have always been fascinated with using mother nature to our advantage in the laboratory, especially when curing cancer is the goal.  When I heard a talk given by Jake at Moffitt during an away rotation for radiation oncology in mid-2015, I knew he was doing some really exciting work that I was interested in doing and that was tangentially similar to work I had done in the past. We spoke briefly after his talk and threw around the idea of me returning during January and February to work in his lab. After meeting with Jake a few times, everything fell into place, and I am on board to start doing research with Jake for the next 2 months.

Although 2 months is not a long time to spend in a wetlab, I am excited to begin a handful of projects that will hopefully produce impactful data. Jake was the first to introduce me to the concept of mapping evolutionary landscapes. A paper published in Science in 2006 showed that there exist certain pathways that populations take in order to evolve to handle a selective pressure (see Once these landscapes are mapped, they can be used to steer populations along certains paths, such as a path that leads cancer cells to be sensitive to a drug (see Figure 1). In order to begin creating these maps, cancer cell lines need to be evolved to gain resistance to various drugs. My active role in this project will be performing radiation sensitivity and genetic assays on these cancer cell lines at different time points during their evolution against various chemotherapy drugs (see Figure 4). The idea is to gain insight into temporal changes cancer cells experience when exposed to chemotherapy drugs and to determine if there are key time points when synergism between chemo and radiation are at its best and its worst.

The next project involves engineering non small cell lung cancer strains with common mutations that confer resistance to chemotherapy drugs to measure evolutionary landscape. Rather than relying on mother nature to create these mutations, I will be utilizing CRISPR/cas9 to perform common genomic edits on non small cell lung cancer strains to create resistance. These cell lines will then be used to infer evolutionary landscapes developed in prior evolutionary experiments and to create mathematical models to predict chemotherapy regimens that minimize resistance.
I remember being asked during my interviews for medical school the question of where I see myself in 5 years. I would answer that I imagined myself as a clinician who is actively involved in research. At the time, I did not know what type of clinician I wanted to be or what role I would play in research, but I knew that both aspects of medicine were necessary to satisfy my insatiable curiosity and my desire to help people directly. I can honestly say as I approach that 5 year mark that I am beginning to see my vision become a reality, and this beautiful marriage between scientist and clinician is more perfect than I could have ever hoped.

Jeff can be found on twitter at: @ggcancer898

Saturday, October 31, 2015

bioRxiv vs. arXiv

If you've been reading along since the beginning, you'll know that I'm a huge #openaccess fan, and, really, am something of an oversharer. When I began my scientific career, I was posting all of my work on the qBio section of the arXiv. Me and some colleagues responded to the increase in utilisation of #preprint servers by making Warburg's Lens, a blog inspired by Haldane's Sieve to help aggregate mathematical oncology pre-prints, and allow for discussion. We've had a ton of success with Warburg's Lens, and it has been helpful for many folks. One of the nice aspects of it, that the arXiv didn't have, was the ability to comment, and to link in to social media in general.

A couple years ago, Richard Sever contacted me about an upcoming project, the bioRxiv, and asked me to be an affiliate - which I eagerly accepted. It seemed the perfect way to help convince my more open minded biological colleagues - something to which there are still quite a few (unsubstantiated) barriers.

This past summer, at the annual Society for Mathematical Biology Meeting, I gave a talk on pre-print servers and social media in science. While I was preparing for the talk, I ran across a great infographic about the arXiv and was blown away how LITTLE qBio there actually is... (something like 1.6% - TONS MORE info here)

Anyways, I recently asked Richard if there were any stats on publications/utilization of the bioRxiv. At the time, there were no stats done, but today I saw on this twitter:

Within just a year, more than half have been published!  Good news. I'm hoping that my newest preprint (which is also on Warburgs Lens here) joins the majority ASAP :)

Anyways, I'm heartened to see the increase in utilisation.  I have to say, having submitted I think 6 manuscripts to the bioRxiv, and around the same number to the arXiv, the process at the bioRxiv is MUCH easier. It can be done in minutes, rather than fighting with the LaTeX compiler on the arXiv istelf. My current practice is to send everything to the bioRxiv, though I still read and monitor the qBio section of the arXiv... what are your practices?