stuck in a bad project

God bless the Zheng Lab and their cheery lab monkeys. As I stare into the growing abyss each day this reminds me that dozens of other poor b******s do the same. God bless our PI’s.

parallel computing – biology style

I’ve been reading Regot et al’s Nature paper, out last November which is very similar to my project. It’s very interesting if anybody is into building biological analogues of electrical circuits. The work by Regot, led by Ricard Solé and Francesc Posas at University Pompeu Fabra in Barcelona has used a distributed computing approach and built logic processing circuits as an emergent property of many cells working together with their own individual simpler circuits. I find this quite interesting as, as the author states, building complex circuits in single cell lines can be a complicated task. Many components must interact predictably and robustly within a complex matrix of pre-existing interactions that are poorly understood. Ron Weiss of MIT presented a workshop at FEBS in Alpbach 2 years ago and explained how he could build and model interactions between a promoter and a reporter gene, but going beyond 2 promoters the behaviour becomes almost completely unpredictable. There’s just too much stochastic noise and underlying lack of understanding of the cellular chassis (hence the push for Craig Venter’s synthetic cell) to build predictable and reliable behaviours that can be thoroughly characterised and remain stable over a prolonged time period. Using very simple interactions in individual cells is simple to build (theoretically!), and provides a library of simple constructs. These libraries can then be hot swapped (to use a computer geek term) with each other to make much more complex behaviours from the culture as a whole. This is a true combination of systems and synthetic biology. The individual circuits remain robust as they provide a lighter load on the cell line to maintain, and the emergent property subsequently remains robust (for >9h in the paper). The combinations can also be simulated and optimum mixtures designed for the desired circuit behaviour.

In systems as complex as yeast (compared with E. coli) this is a much more desirable framework to work, in my humble experience anyway. Libraries of strains can be collected and combined to explore complex behaviours or engineer interacting microbiomes with useful engineering applications like biosensing and biological computing. Previous papers in Nature to do with bio-computing have utilized RNA aptamers (e.g Win and Smolke, 2008) to enable switching on and off gene expression to replicate boolean logic, however RNA secondary structure can be difficult to predict, and can form many different conformations under different physiological conditions. Designing RNA aptamers can also be challenging, compared with cloning. One could imagine this leads to RNA aptamer structures being unstable or the results difficult to replicate across laboratories without highly skilled expertise. It is interesting to compare this kind of approach with using multi-cellular systems for biological computing.

I wonder if the type of systems being built by Ragot et al could combine eukaryotic and prokaryotic systems and exploit features of each, such as the fast growth of E. coli with the signal processing abilities of yeast. One could exploit the environmental robustness of Saccharomyces or Candida with motile prokaryotes etc. I imagine the growth cultures and the potential for each to attempt to out compete the other would be somewhat complex!

pressure

Still no data from the lab, and many broken plasmids to sort out. Just ordered some additional components that will need to be investigated. They should arrive next week.

There is less than 10 months remaining to submission and many of my colleagues are sorting out their final experiments before beginning to write up. There’s no denying this project has been a disaster from start to finish. The months of lost time during the lab move compounded the problem, but I’m hoping to get something out of the new qPCR machine, couple it with some westerns, throw it in a model and get the hell out of dodge. I’m not sure if any Ph.D student has come this close to the wire, but it’s certainly an exercise in mental stability. If I can generate data by end of February then it should be a good stress reliever. Game on, I guess….

qPCR

After having a look around at the available instruments for qPCR the lab has invested in a Qiagen Rotor Gene Q. This should enable me to do accurate quantification of mRNA from the gene circuit without having to do northern blotting. I am hoping it will fast forward the data generation to move into modelling. Training is booked in for tomorrow, so we will see what it can do…

one small step for your PI, one giant leap for you

The lab has moved to it’s new home at University of Warwick. After a few months of packing, unpacking, packing, unpacking, packing again, and then unpacking, our stuff is out on the benches and the lab is starting to function. The project can continue after 3 months of effective down time. How this will effect the final year of this work I have no idea yet. I was behind before I started and I wonder if I’ve already reached critical mass anyway.

The new lab is functional. Not quite the state of the art MIB but the necessary facilities are available. The University has a different culture to Manchester and will require some acclimatization. The University is undergoing some “restructuring” in these “hard economic times”, and it’s not hard to see the University is somewhat financially challenged.

On the upside we have a nanodrop (don’t leave your lab without one!) and in a rather unexpected turn of events we can invest in a qPCR machine. I have been investigating the Roche Lightcycle 480, Applied Biosystems StepOne and a BioRad module for our c1000 thermal cycler. The Roche machine doesn’t need a ROX standard for normalizing the signal from the heating block and is interchangeable between 96 and 384 well configurations. Separate filter sets for the laser also gives some future proofing and flexibility. Will see what the StepOne looks like later this week.

On a lighter note, my colleague Natalie Stanford over at PhD Fodder.co.uk posted this hilarious article about Anatidaephobia – the fear of being watched by a duck. An epic fail for the site’s advertising robot. Anatidaephobia is a worrying condition on a campus that boasts in the prospectus of having more ducks than students. Who’s watching who’s watching who?

back home

Back in Manchester after a 20 hour marathon across Europe from Spetses, and working through the depression of leaving a Greek island for cold, damp England. Many things to do now that the final year has arrived. Spetses conference was useful in providing some contacts that can provide additional yeast strains for me to work with and some important changes I can make to the modelling side of my project. Much thanks must go to the biochemical society for providing the travel and conference funds that it possible to attend.

I will change my modelling from mass action kinetics to more michaelis menten style with hill coefficients. Mass action kinetics can’t saturate but at some point the transcription and translation reactions will. The DNA and mRNA transcripts will at some point max out the number of polymerase and ribosome molecules. I can also incorporate expressions for activation and inhibition which will be important to model correctly in order to simulate the feed forward loop. The equations will be derived from Professor Pedro Mendes’ 2003 paper.

Final lecture

Quick swim then back for final lecture from yoshi nakamura of tokyo. RNA bio mimicry for designing therapeutic drugs from aptamers. Interesting animal model results for treating auto immune disease. Some cash issues for scale up to kg of rna however.

Igem workshop this morning. We had 2 hours to work up a biological engineering project from concept to project proposal. Many fun ideas were presented. Good team working exercise. Later phil hollinger presented his work on protein evolution. Interesting research on possible origin of self replicating molecules. Moved onto synthetic information carrying molecules and potential routes for building orthogonal circuits.