Posts Tagged ‘systems biology’

No hypothetical vaccines please!

14 January, 2011

A new editorial in Elsevier’s Vaccine, by Gregory Poland and JR Hollingworth, gives one much food for thought…especially if one and one’s associates are engaged in vaccinology, however quixotic that quest may be.

Especially quixotic when certain editors take 11 months not to publish one’s HIV vaccine paper, but that’s a story for another day…!

The article is entitled “From Science II to Vaccinology II: A new epistemology“, and is a thoughtful and quite intellectually challenging piece of work.

I have previously indicated that I am not a fan of hypothesis-driven science, however well entrenched it is in the psyches of most who practice it – in fact, I have gone as far as claiming elsewhere (thanks, Alan C!):

“Profound Insight No. 1: hypotheses are the refuge of the linear-thinking.

…I am quite serious in disliking hypothesis-driven science: I think it is a irredeemably reductionist approach, which does not easily allow for Big Picture overviews, and which closes out many promising avenues of investigation or even of thought. And I teach people how to formulate them so they can get grants and publications in later life, but I still think HDS is a tyranny that should be actively subverted wherever possible.”

And here we have two eminent scientists agreeing with me!  Not that they know that they are (or care, I am sure), and nor is it important – for what they have done is write a tight and carefully reasoned justification for moving away from the classical approach in vaccinology, as the complexities of the immune system and responses to pathogens and vaccines render the reductionist approach inadequate to address the problems at hand, and especially those presented by rapidly-mutating viruses.

This really is quite a profound suggestion for change, as the world of vaccinology is notoriously conservative, and it is really difficult to get people even to discuss only mildly paradigm-nudging concepts – oh, like cellular responses possibly being as important for protection against papillomaviruses as sterilising antibody responses? – let alone publish them.

Their final paragraph is especially apposite:

As we move into the world of Vaccinology II, or the “second golden age of vaccinology”, success will come only with the willingness to minimize the current Newtonian framework of thinking, and to adapt a new framework (Science II) that requires novel advanced bioinformatic and chaos theory-like analytic approaches, as well as multi-level systems biology approaches to studying currently unpredictable and uncertain self-organizing complex systems such as host immune response generation. Such work is difficult, expensive, challenging, and absolutely necessary if major advances are to occur in vaccine biology generally, and vaccine immunogenetics specifically.

This is fundamental stuff: I sincerely hope people in the field of HIV vaccines in particular give it some heed, as there the funding paradigm has actually shifted back towards requiring that everything be “hypothesis-driven”  – and I think this is a retrograde step, when the funding agencies (NIH, Gates Foundation) need to take more, rather than fewer risks, if we are to make any meaningful progress in our lifetimes.

While I am also not a fan of “systems biology” – because I think it is a catch-all term for what amounts to multidisciplinary research, and many of its proponents are brash snake-oil salesmen – modern vaccinology  really is a fertile field to plough using the new approaches.  Poland and Hollingworth put it well:

Similarly, as applied to understanding host variations as causative of inter-individual heterogeneity in immune responses  to such viruses, a Newtonian–Descartian view is entirely inadequate….

Rather than general principles, Vaccinology II and the new biology  is increasingly informed by principles such as pattern recognition, systems with non-linear qualities, and complex networks—often  focused at the individual, rather than population, level.

Amen to that.  Now, to get some money to do that…!!  B-)

Venter can do WHAT for influenza??

5 June, 2010

I have kept out of commenting on what J Craig Venter and others have done recently, given that many others have done so, and done so well – however, there is recurring mention of what “this technology” could do for influenza vaccines specifically, which has both puzzled and intrigued me, given a distinct lack of obviousness.

So I will comment, if only to clarify this issue for me and anyone else who cares.

To give some background, the New Scientist issue of the 29th of May has a guest editorial by J Craig Venter, Clyde Hutchison and Hamilton Smith, where they discuss some of the implications of their having made a totally synthetic and viable Mycoplasma mycoides genome (see also Science, DOI: 10.1126/science.1190719).

So what, exactly, is it they did?  OK, so they spent US$40 million or so constructing a genome, in segments, from sequence information housed in an electronic database, via chemical synthesis of long stretches of DNA.  They then assembled these segments into a singular genome in yeast, and then inserted this into cells of the closely-related Mycoplasma capricolum which had been stripped of their genomes – and incidentally, rendered unable to destroy the incoming genome as “foreign”, by a process which is now proprietary.  These cells then expressed the new DNA, which allowed them to multiply, and to take on all of the characteristics of the synthetic M mycoides, given that all of the original cell constituents from the original bug (proteins, mRNA, membranes, etc) would be turned over in time, and become those specified by the new genomes.

This is a big deal – a really, really big deal – but at the same time, they themselves recognise it is an incremental step in a long series of steps that started with Arthur Kornberg’s lab making the first complete synthetic and viable genome of a virus (phiX174) by in vitro synthesis from virion DNA, polymerase and nucleotides.  In fact they modestly point out that this is not even the first  complete cell genome that has been synthesised; it is the largest, however, and the one that worked.  They were not so modest in missing out a few other landmarks before their own complete synthesis of phiX174 in 2003, however: for example, the first synthesis of a functional plasmid, and the first generation of a RNA virus genome from a cDNA copy, and the complete synthesis of an infectious poliovirus genome, are not mentioned.

So what is it they did not do?

Well, they did not “create life”, however much even relatively respectable publications might claim they did: life is a lot more complex than chemicals, and people have “rebooted” cells before with exogenous genomes; what they did is not really qualitatively different to infecting a cell with a synthetic virus.

They have also not done anything that is immediately useful: their new organism differs from the original only in having a few genes missing, and a long literary message and ownership-encoding “watermark” inserted.

More positively, they have also most emphatically not opened the floodgates for bioterrorists to mail order complete poxvirus or anthrax genomes: as I have noted here previously,

“…There are more than enough nasty agents out there that are relatively easy to obtain, and do simple kitchen-based microbiology with, to keep entire cave complexes and Montana libertarian enclaves busy for years, without resorting to complicated molecular biology”.

Or spending $40 million dollars.  And I will say it again….

So aside from the details, what have they done?  In the NS editorial, Venter et al. say this:

“We now have the means to design and build a cell that will define the minimal set of instructions necessary for life, and to begin the design of cells with commercial potential, such as fuel production from carbon dioxide. We can assemble genome-sized stretches of DNA that can also be used to mix and match natural and synthetic pieces to make genomes with new capabilities.

Synthesising DNA in this way is still expensive, but we expect the cost to fall dramatically. This may make the complete synthesis of genomes competitive with the alteration of natural genomes to add new capabilities to bacterial cells. It should also be practical to synthesise simple eukaryotes, such as yeast, to which it is already possible to add extra chromosomes. The construction of large pieces of synthetic DNA and their introduction into a receptive cytoplasm is no longer a barrier. The limits to progress in synthetic biology are now set by our ability to design genomes with particular properties.”

Right: so what they have done is set the benchmark for what is possible – rather than what should be done.

Because it is a lot easier to do things such as they propose by other ways – as is pointed out in the companion article to the NS editorial.  For that matter, I am sure one could more easily end up with a completely synthetic and much larger cellular genome by incrementally replacing genome chunks by homologous recombination or transposon-mediated insertion / Cre-Lox deletion, and have it cost far less and be less subject to error, than by synthesising it de novo.

Influenza virus - Copyright Russell Kightley Media

And how does any of this relate to influenza vaccines?

The only comment I can find in the NS article that sheds light on this is the comment:

“As soon as next year, the flu vaccine you get could be made synthetically,” he [Venter] says.

Except that this has been possible for years already, after the poliovirus synthesis…?  I think it was rather a badly-chosen example rather than any actual plan; however, there is not a lot of point in making a synthetic influenza virus genome, given that the attenuated originals already work quite well as vaccines – and we don’t yet understand how to specify avirulence in influenza, so any synthetic version would necessarily be a copy of an extant version.

So hype rather than fact for ‘flu; promise rather than substance for carbon dioxide sequestration and biofuels – but still the coolest thing since sequencing your own dog…B-)