Infocom, HHGG, and a blast from the past.

For anyone who grew up as a computer geek in the 1980s, this blog post from Andy Baio will be of very great interest. It seems that he has come across the network drive from Infocom, the premier interactive fiction company of that time, and that one of the things on the drive is a fragment of an unreleased sequel to the Hitchhiker's Guide to the Galaxy game. There's not much to the fragment itself, but the comments on the site and the emails that he quotes are priceless.

Enjoy. (Link)

Last Day: Liveblogging the Molecular Evolution/Infectious Diseases Keystone Conference

Day 3
Eun-Chung Park (NIAID) made an announcement about NIH funding opportunities for non-US investigators. There is a new PA for international research in infectious diseases PAR-08-130.

Morning Session
David Alland (UMDNJ) gave a fantastic talk on the evolution of drug resistance in M. tuberculosis. He uses SNP-based phylogenetic analysis to link specific mutations to drug resistance phenotypes. In the setting of confusing data as to which mutations actually cause resistance and which are simply epidemiologically linked, he showed that phylogenetic analyses can provide answers, including answers that impact patient care, that other techniques cannot.

M. Cristina Gutierrez spoke about “rethinking tuberculosis in the light of evolution.” She focused on smooth-colony variants of Mycobacterium tuberculosis (MTb), which are an uncommon but confusing cause of human disease. Using MLST, she showed that MTb can be divided into distinct lineages, which she calls MTb and Mycobacterium prototuberculosis. She provided estimates of the age of the split between the two lineages and hypothesized about the relevance of this to TB in hominid ancestors.

Bruce Levin delivered an “opinionated rant” about recombination in bacteria with a focus on transformation. Using a combination of equations, experiments, and “yak-yak,” he put forward a new hypothesis about the evolution of transformation and genetic competence. He hypothesizes that although competent bacteria may have a growth deficit in comparison to non-competent competitors, in the setting of occasional, punctuated selective pressure, the competent strains will predominate.

A short talk describing MLST analysis was given by Sebastien Gagneux from MRC. He focuses on the biogeography of MTb and links phylogenetic analysis to hypotheses about human migrations.

Anne Buboltz from Penn State gave a short talk about a strain of Bordetella bronchiseptica that has become less virulent due to a recombination event. Her group mapped this recombination, and using careful genetic studies linked the genes involved to the loss of virulence.

In the afternoon session, Jim Musser gave a phenomenal talk about his group’s integrated systems biology approach to group A streptococcus virulence. Using longitudinally collected human disease isolates, full genome sequencing, and unique animal models, they have linked single mutations on the streptococcal chromosome (including a single nucleotide change in one case) to both overall virulence and (in at least one case) to anatomic site specificity. Amazing, amazing work. I plan to go back and read several of the papers that he described again.

Nancy Moran gave an exciting talk about the evolutionary genomics of bacterial symbionts of insects. This is an area that has exploded recently, especially with the publication of the paper taking a genomic look at the dual sharpshooter symbiosis. She gave an overview of the field, focusing on the common characteristics of genomes of obligate symbionts (small size, loss of metabolic functions), the importance of maternal passage of symbionts to their evolution, and how more recently established symbioses may differ from ancient ones.

Mark Achtman (University College Cork) spoke about the complex and intimate relationship between Helicobacter pylori and humans. In addition to learning a tremendous amount about the organism itself, his group has used the rapid evolution and recombination within H. pylori to trace human migrations at a level of resolution that human genomic studies cannot approach. This is an extremely useful and creative use of evolutionary and phylogenetic analysis and a field that is just beginning.

Martin Antonio spoke about Streptococcus pneumoniae, an extremely important cause of childhood mortality worldwide. His group looked at invasive S. pneumoniae disease in Africa and, through a detailed MLST analysis, linked a specific serotype to severity of disease.

Day 2: Liveblogging the Molecular Evolution/Infectious Diseases Keystone Conference

Day 2
Morning Session: DNA viruses
Elliot Lefkowitz gave the opening session today. He spoke about gene acquisition and loss as driving factors in speciation of poxviruses. He focused on the orthopoxviridae (the group containing vaccinia, variola, monkeypox, and other pathogens). They have developed computational tools to do gene prediction and annotation across all known poxvirus genomes. This has allowed them to get around problems with inconsistency in annotation among deposited sequences. Gene acquisition in the core region appears to play much less of a role in orthopoxvirus speciation. These techniques have also enabled them to look at the process of gene fragmentation and loss over the evolutionary history of these viruses. They have begun to test these predictions, especially with regard to promoters. Some of this work is in a recent PNAS paper:
Assarsson, E., Greenbaum, J.A., Sundstrom, M., Schaffer, L., Hammond, J.A., Pasquetto, V., Oseroff, C., Hendrickson, R.C., Lefkowitz, E.J., Tscharke, D.C., Sidney, J., Grey, H.M., Head, S.R., Peters, B., Sette, A. (2008). Kinetic analysis of a complete poxvirus transcriptome reveals an immediate-early class of genes. Proceedings of the National Academy of Sciences, 105(6), 2140-2145. DOI: 10.1073/pnas.0711573105

The parvovirus family was the subject of the next talk, by Colin Parrish from Cornell. His title was “How to change your host range in a few easy steps: canine and feline parvoviruses.” In the 1970s, there was a host range shift within a feline parvovirus (FPV) that allowed it to infect dogs rather than cats. Within a year or two, this (now) canine parvovirus (CPV-2) spread through a wide geographic area and shifted again to allow efficient spread to both cats and dogs. All of the canine parvoviruses today appear to be descendants of this virus, and most are in the CPV-2a group. Nearly whole genome analysis has allowed description of geographically distinct clades within the CPV-2a group. Certain residues on the threefold spike of the capsid protein appear to control host range. The host receptor is the transferrin receptor, and binding to it triggers chathrin-mediated uptake.

Thomas Schultz talked about diversity within the rhadinoviruses (RV, which are gamma-2-herpesviruses, including KSHV – the causative agent of Kaposi’s sarcoma). There are two major branches within the RV lineage, and these are distant relatives of Epstein-Barr virus (EBV). Within KSHV, there was an early split into a number of geographically distinct lineages. More recently, there has been significant recombination among these lineages as well as introduction of sequences at the ends of the genome. KSHV protein K1, which contains hypervariable regions and has oncogenic properties, is located at one end of the genome at a position analogous to LMP-1 in EBV. KSHV protein K15, which activates host cell proinflammatory signaling, is at the other end. Homologues of K15 appear in other related viruses and seem to have distinct signaling profiles.

Mary Odom from the Lefkowitz lab gave a short talk about gene acquisition by poxviruses via horizontal transfer. She has developed bioinformatic techniques to look for putative transfer events. By looking at multiple databases (viral vs. eukaryotic or viral vs. bacterial), they can generate 2D plots of best BLASTP scores to generate leads.

Edgar Sevilla-Reyes gave a short talk about the dynamic gene content of human cytomegalovirus (CMV). Gene disrupting mutations occur after only a few passages in vitro. After more passages, larger deletions predominate, especially at the ends of the UL sequence. They studied genetic integrity at the ends of UL for 22 CMV isolates and found striking clustering of the sites of mutations. These same sites were also found to be mutated in a subset of non-passaged clinical CMV isolates, raising the question of what the selective pressures on these sites might be.


Workshop
short talk session

Nels Elde described an evolutionary arms race between poxviruses and their hosts. Protein kinase R (PKR) is a host factor with antiviral effects. In its dimeric (active) form, PKR phosphorylates eIF2alpha, which leads to a global transcriptional arrest. This blocks viral protein synthesis and can aid in host defense. As evidence of the importance of this strategy, several pathogenic viruses encode substrate mimics to block PKR. These mimics (K3L is the poxvirus example used here) are structurally similar to eIF2aplha and can block the PKR/eIF2alpha interaction. Analysis of the dN/dS ratio suggested that PKR was under positive selection but eIF2alpha was subject to purifying selection (i.e. PKR is undergoing adaptive evolution, while changes in eIF2alpha are generally selected against). By mapping specific sites in all three of the involved proteins (PKR, eIF2alpha, and K3L), the picture of a rapidly evolving selective race appears.

Willem van Ballegooijen talked about surveillance of hepatitis B virus (HBV) sequences following targeted vaccination of risk groups (not global vaccination!) in the Netherlands, starting in 1998. He used coalescent-based analysis of HBV DNA sequence and Bayesian analysis to provide statistical support for a decreasing incidence of infection, starting in about 2000.

Intrahost diversity and evolutionary dynamics of influenza were the topics of
Elodie Ghedin’s talk. Because of evidence of reassortment among cocirculating clades of influenza during a single season (see yesterday’s talks), she hypothesized that coinfection of individuals must occur. Using samples from human isolates, she examined diversity among influenza sequences. She found convincing evidence for double and even triple infections with influenza subtypes (which is something that we see clinically as well) and hypothesizes that this may be a source for flu reassortment and diversity during the active season.

Ma Luo from the University of Mannitoba discussed evolutionary interactions between HLA alleles and HIV-1 infection within an East African sex worker population. She studied women from the Pumwani sexworker cohort (high prevalence) and a mother-child cohort (low prevalence). They examined linkage of HLA class I alleles to the risk of seroconversion during the study and grouped these into “protective” and “susceptible” alleles. Over time (1985-1992 vs. 1993-2001) there was a decrease in the susceptible allele prevalence and an increase in the protective allele prevalence in the high-risk cohort.

Molly OhAinle talked about the evolution of the APOBEC3 locus of anti-retroviral elements in hominids. She presented evidence that the anti-retroelement of APOBEC3H, which is a broadly active member of the family, has undergone very recent changes. Macaque APOBEC3H efficiently inhibits HIV infectivity, which human APOBEC3H does not. The loss of this activity of APOBEC3H appears to have occurred independently at least twice in recent human evolution. Human APOBEC3H is considerably less stable than that of other primates, which accounts for its lack of activity. Ancestral sequence reconstitution suggests that recent human ancestors had stable (effective) APOBEC3H. Is there selection for non-functional APOBEC3H alleles in the human population?

Bouke de Jong (NYU) examined the phylogeography of the Mycobacterium tuberculosis complex, specifically the appearance of M. africanum in West Africa. By looking at genotypes of organisms from smear-positive tuberculosis (TB) cases and their contacts. A significant portion of TB in the Gambia appears to be caused by M. africanum. There are differences in risk of progression but not transmission or severity when comparing M. africanum with M. tuberculosis.

Joel Wertheim used Bayesian dating strategies to calibrate a molecular clock for SIV evolution. They used envelope and polymerase sequences to develop their model, which allowed specific estimation of dates for divergence of SIV clades as well as HIV-2.

Tsan Yuk Lam discussed transmission dynamics of natural reassortant H5N1 influenza in Indonesia. His analysis is consistent with a single clade arising from a single H5N1 introduction from China. There has been intralineage reassortment within Indonesia over a short time.

Evening Session: Retroviruses

Beatrice Hahn (UAB): Origins of HIV-1 and adaptation to its human host
John Moran (U. Michigan): Studies of a human transposable element (LINE elements)
Nathan Wolfe (UCLA): Cross-species transmission and establishment of novel human retroviruses.
Christine Clouser (U. Minnesota): Short talk: Exploiting the mutation rate of HIV

Day 1: Liveblogging the Molecular Evolution/Infectious Diseases Keystone Conference

Day 1

Today’s talks were focused on the evolutionary dynamics of RNA viruses, including influenza, dengue, measles, and SARS coronavirus.

Morning session:

Jeffrey Taubenberger led off the day by talking about influenza, the “continuously emerging infectious disease.” He emphasized the importance of the explosion of sequenced influenza genomes that have become available in the past several years. Starting with a 2005 PLoS Biology paper that looked at H3N2 influenza viruses, he discussed the process by which a minor influenza clade can transition into a dominant clade. A recent PLoS Pathogens paper used concatenated genomes to generate a phylogenetic tree. This analysis showed that there is little antigenic drift during the influenza season – drift seems to occur primarily between seasonal peaks of flu. This leads to a “source-sink” model of the generation of influenza diversity: sites in the tropics may serve as a supply of genetic diversity that leads to differences in seasonal strains elsewhere. Also, there is tremendous influenza sequence diversity within wild birds, and within host genetic transfer is thought to generate a large pool of reassortant viruses. A host switch to humans leads to a dropoff in diversity, thought to be due to lack of access to this tremendous sequence pool. The selective pressures on influenza viruses within avian hosts are only beginning to be explored.

Holmes, E.C., Ghedin, E., Miller, N., Taylor, J., Bao, Y., St. George, K., Grenfell, B.T., Salzberg, S.L., Fraser, C.M., Lipman, D.J., Taubenberger, J.K. (2005). Whole-Genome Analysis of Human Influenza A Virus Reveals Multiple Persistent Lineages and Reassortment among Recent H3N2 Viruses. PLoS Biology, 3(9), e300. DOI: 10.1371/journal.pbio.0030300

Nelson, M.I., Simonsen, L., Viboud, C., Miller, M.A., Holmes, E.C. (2007). Phylogenetic Analysis Reveals the Global Migration of Seasonal Influenza A Viruses. PLoS Pathogens, 3(9), e131. DOI: 10.1371/journal.ppat.0030131


Edward Holmes talked about the molecular evolution and phylodynamics of dengue virus. The 4 serotypes of dengue (DENV-1 through -4) seem to have split ~2000 years ago, with most diversity appearing in the past several hundred years. Studies of the microevolution of dengue in Bangkok showed complex interactions among serotypes, with DENV-1 through -3 largely in phase and DENV-4 predominating when other serotypes were less prevalent. There are also distinct genotypes within each serotype, with DENV-2 undergoing significant genotype changes in Vietnam between 1999-2007. Importantly, as we move toward phase 3 trials of investigational dengue vaccines, the combination of detailed spatial data and dengue genome sequences will be indispensable for understanding the consequences of vaccination.

Oliver Pybus from Oxford presented new data about the complex within-host evolution of hepatitis C virus that occurs during chronic infection. Using longitudinal data from a number of chronically infected patients, they derived multiple viral sequences over several years. They found marked diversity over time, considerably more than was seen in a similar longitudinal study of HIV. This was thought to be consistent with discontinuous variation – i.e. low-level variation of HCV punctuated by periods of vastly increased diversity. This may be the result of a sequestered hepatic pool of HCV genetic diversity, which seeds the plasma at different times.

Paul Rota from CDC gave an overview of the global laboratory-based surveillance for measles and mumps. This talk was particularly important in light of the ongoing outbreak of measles in Switzerland that has led to a number of imported cases in the U.S. (the index case was, not surprisingly, an unvaccinated child). Genotyping of measles viruses worldwide has greatly enhanced our understanding of its epidemiology. There is a reasonable amount of geographic restriction of measles genotypes. Importantly, though, genotyping can help trace the origin of imported cases. Despite the presence of multiple genotypes, measles is still a monotypic virus, and the vaccine (which is a genotype A virus) provides protection across all measles genotypes. Likewise with mumps, where vaccination with the Jeryl Lynn strain continues to provide protection across all circulating mumps strains.

These were followed by two short talks. The first was by Phillipe Lerney, who uses Bayesian models to infer probable modes of geographic spread of a variety of viral diseases. Naomi Forrester spoke about rabbit haemorrhagic disease virus. Her group did a phylogenetic analysis of circulating RHD sequences in healthy rabbits in the UK. There were 2 major circulating subtypes, an epidemic strain and a “widely divergent” strain, and they hypothesize that there may be some degree of cross-protection between these two clades.


In the evening session, Kathryn Holmes gave a great talk on coronaviruses (CoV) and host adaptation. The groups of CoV (1, 2a, 3, and 3b) each contain tremendous diversity with respect to host range and receptor specificities. The spike glycoprotein (through its S1 region) dictates receptor binding, but CoV accessory genes are important in host tropism as well. However, many of the accessory genes have unknown origins and functions. There is little homology in these accessory genes among groups, and while they are generally not necessary for in vitro replication, they can have important effects on fitness in vivo. She finished by using mouse hepatitis virus and bovine CoV as models to ask whether viruses can drive the evolution of host receptors.

Alan Barrett from UTMB traced the global spread of west nile virus (WNV) and the incredible diversity of hosts (>300 species of birds, >60 species of mosquitoes) that it can utilize. Sequence analysis showed the presence of 3 distinct clades of WNV (the old world, Eastern U.S., and North American clades), and phenotype analysis of these is ongoing.

Michael Katze gave a description of his systems biology-based investigations of host response to influenza infection. Both reconstructed 1918 and H5 flu are highly virulent in murine and primate models of infection. Microarray-based profiling of host responses showed an important role of control of the early inflammatory response in the lung for survival. Animals infected with 1918 flu had a rapid and uncontrolled response that led to death.

The final talk of the day was from Chun-Chau Hon, who did a phylogenetic analysis of coronaviruses and a Bayesian molecular clock analysis. By examining a novel clade of bat SARS-like CoV, he made predictions regarding the potential ancestor of human SARS CoV. Some of this work was recently in J Virol.

Hon, C., Lam, T., Shi, Z., Drummond, A.J., Yip, C., Zeng, F., Lam, P., Leung, F.C. (2007). Evidence of the Recombinant Origin of a Bat Severe Acute Respiratory Syndrome (SARS)-Like Coronavirus and Its Implications on the Direct Ancestor of SARS Coronavirus. Journal of Virology, 82(4), 1819-1826. DOI: 10.1128/JVI.01926-07

Liveblogging the Molecular Evolution/Infectious Diseases Keystone Conference

I am in Breckenridge at the Keystone Symposium on Molecular Evolution as a Driving Force in Infectious Diseases. This is my first attempt at liveblogging a conference, and I'll do my best to post highlights of each session. My training is in infectious diseases, not evolutionary biology, but I'm here because our most recent paper, which described a new member of the cholesterol-dependent cytolysin family of bacterial toxins, led me into a project looking at the evolutionary history of that family. I'll be presenting some of that work later in the conference and am hoping to learn from people who think about these things for a living.

So far, the conference has been great. It is a very internationally representative group, something that Keystone has been striving for with their Global Health Series of symposia. The evening session today was a plenary lecture by Sir David Weatherall, who spoke about the interaction between malaria and human evolution. The title of the talk was "The Role of Genetic Variation in Susceptibility to Malaria in Human Evolution: Was Haldane Right?" He started off with brief descriptions of the pathophysiology of malaria and the hemoglobinopathies. He described JRS Haldane's response to a description of thalassemia mutation rates at a conference in 1948, in which he postulated a heterozygote advantage for carriers of hemoglobinopathy alleles. Weatherall reviewed his data, collected over decades, and those of other groups, showing how far we have come toward a mechanistic understanding of the interaction between hemoglobinopathies and malaria and how powerful the selective pressure of some diseases can be on the shaping of human evolution. Two issues that I had never even considered were the interactions between alleles in people who carry multiple mutations (such as HbS trait/alpha-thalassemia or HbE/beta-thalassemia) on overall malaria susceptibility (overall point: it can be quite complicated, and some of these interactions may even restore wild-type susceptibility), and the possible implications of these very frequent alleles as we move forward with malaria vaccine trials in Africa and elsewhere. Some of the most recent work on polygenic interactions is covered in a recent PNAS paper as well.

I'm looking forward to the sessions tomorrow, which focus on viral diseases, and I'm also hoping that FedEx delivers my poster as scheduled...



O'Donnell, A., Premawardhena, A., Arambepola, M., Allen, S.J., Peto, T.E., Fisher, C.A., Rees, D.C., Olivieri, N.F., Weatherall, D.J. (2007). Age-related changes in adaptation to severe anemia in childhood in developing countries. Proceedings of the National Academy of Sciences, 104(22), 9440-9444. DOI: 10.1073/pnas.0703424104

A good day for open access

The NIH public access mandate goes into effect today. Jonathan Eisen has a good post about this (and, more specifically, about the role of the Public Library of Science in making this happen) at The Tree of Life. Importantly, a growing number of journals are making compliance with the NIH mandate automatic, and those that have not yet will be under increasing pressure to do so as we go forward.

JBC offers an open access option

In a welcome move, the Journal of Biological Chemistry has announced that in addition to their free-access papers in press feature, they will now allow authors to elect (for a fee) an "Author's Choice Publication Option." Essentially this allows authors to have their work immediately available under a Creative Commons Attribution Non-Commercial license (not quite as open as the PLoS Journals, but certainly not bad). It appears in their instructions to authors that copyright is still transferred to ASBMB for all manuscripts, though.

I've published some of the work that I am most proud of in JBC, and it is great to see them making open access an option for authors.

Measles outbreak in U.S. started by unvaccinated child.

The measles outbreak that started in San Diego was imported by an unvaccinated 7 year old boy who visited Switzerland. The MMWR early release report is here. There have been at least 11 secondary cases, and a large-scale public health response has been required to deal with the many contacts of this child who are also unvaccinated because their parents used "personal beliefs exemptions." How many of these does it take before we do away with these ridiculous exemptions? Parents who refuse vaccines not only put their own children at risk, but also risk the lives of children who have to rely on herd immunity for protection (immunocompromised children who cannot receive the live-attenuated measles vaccine and those too young to be vaccinated).

Silver Guy, San Diego, 3/29/2008