Equinox, or the two times a year I actually think about science.
Top 12 Scientific Questions
These are the top scientific questions I am currently fascinated by, would like to see a conclusive answer to in my lifetime, or believe striving to address would be fruitful
TCR-pMHC Mappings
Given a TCR, can we reliably and exhaustively predict the set of peptide:MHC complexes
it binds to, and vice versa?
Synaptic Encoding
Can we design a simple experiment to test how a synapse encodes a number?
TCRs vs CAR-T
How can we best translate
TCR-T cells and CAR-T cells in cancers currently refractory to immunotherapy?
Immune Training
Can we design attenuated versions of microbes that selectively conserve beneficial immune training while removing pathogenic effects?
Rational Microbiota
Can we design a standardised cocktail of microbes that recapitulates naturalistic exposure, and administer this to lab animals to maximise translational relevance of research in immunology?
DNA Encryption
Can we robustly encrypt information in DNA in a way that is amenable to large scale training of ML models?
Immune Memory
To what extent is the innate immune system
a mediator of durably protective responses from prophylactic vaccinations?
Phage Therapy
How can we optimise safety and efficacy
of lytic phage therapy against multi drug resistant bacterial infections?
Memory Networks
Can we model the systems-level immune memory of antigen exposure as a neural network?
Memory
What is an ideal memory system?
Is the ideal memory system of humans one that maximises fidelity?
Effector Killing
Can we effectively harness
non-apoptotic mechanisms of cell deaths in cancer immunotherapies?
Virome Legacy
Can we quantify the "fitness costs"
of the virome on human health?
How significant are the positive imprints of the virome, such as priming immunity and protecting against super-infection?
There is no such thing as a pathogen
​
To declare that a certain agent is a "pathogen" is almost always an oversimplification.
Pathogenicity is instead a gradual and contextual phenomenon, that depends on the interplay of a number of both host and microbial conditions. In a given anatomical niche, such as the nasopharynx, Neisseria meningitides is a commensal, although invasion into the bloodstream can lead to severe disease including meningitis. In single infection, S.pneumoniae can be effectively controlled, although concurrent or previous presence of other infectious agents such as Influenza A Virus (IAV) increases susceptibility to pathogenic lung infection. Past interactions of the host with other pathogens can either protect against subsequent exposure, the underlying process exploited by vaccination, or can occasionally worsen response, as when previous antibodies against Dengue Virus (DENV) cause antibody-dependent enhancement of infection by subsequent antigenically distinct strains. Similarly, Mycobacterium tuberculosis infection is typically only "revealed" in aberrant host states of malnutrition or immunodeficiency, commonly induced by AIDS, or lymphodepletion for surgical treatment.
​
The conceptual shift in seeing microbes as potential symbiotic agents instead of unwanted invaders is already well underway, best exemplified by the blooming field of microbiome research. There is an extensive literature on beneficial effects of bacteria on host immunity via xenobiotic metabolism, immune development and colonization resistance. This enthusiasm has mostly been restricted to bacteria, while viral, fungal or other eukaryotic members of the microbiota have not been given so much "generosity of interpretation". Why is the same not true for viruses for example?
​
There are some valid reasons. For example, absence of a universal viral marker equivalent to 16S rRNA for phylogenetic classification, challenges of metagenomic annotation due to considerable sequence diversity in the viral kingdom and the paucity of broadly-acting antivirals to observe systemic effects of viral depletion. One part of the discrepancy might also simply reflect conceptual bias lingering about viruses necessarily being bad (remember that virus means poison in latin). Granted, there are viruses like Rabies viruses that cause 100% fatality, and it would be harder to make the case that they provide net fitness gain to their individual host. But these are the exception and not the norm.
​
One particularly interesting example is the beneficial effects of herpesviral latency on bacterial superinfection. In 2007, Barton et al reported a surprising beneficial effect of latent infection by murine herpesviruses homologous to human Epstein-Barr Virus (EBV) and cytomegalovirus (CMV), which protected mice against secondary bacterial infection by Listeria monocytogenes and Yersinia pestis (yes, they used the bubonic plague as a model!). Using a latency-deficient herpesvirus strain, the authors showed that protection was specifically mediated by latent rather than lytic infection, and identified enhanced bactericidal activity of macrophages and elevated serum IFN-γ as underlying mechanisms. This paper convincingly showed that innate immune enhancement associated with chronic herpesviral infections can provide durable cross-protection against heterologous pathogens.
​
There are several other examples of beneficial effects of viruses in co-infections at the both the single cell (Newcastle Disease Virus) and systemic (LCMV and Vaccinia Virus) levels.
​
I believe that we can dissect the beneficial imprint left by the virome and harness this rationally to enhance immunity. Viruses are not just pathogens, but perhaps a pathogen is not something an agent can "be" anyway.
​
Solzhenitsyn writes in the Gulag Archipelago.
The line separating good and evil passes not through states, nor between classes, nor between political parties either but right through every human heart and through all human hearts
​
From this, I update it to :
​
The line separating beneficial and detrimental runs right through every human virus :)
​
Should we anthropomorphise biology?
​
When writing essays for undergraduate cell biology, one of our tutors used to always slash sentences that attributed any form of agency to biology with a stroke and write: reformulate. No, a virus does not attempt to dock onto a cell. The virus doesn’t want anything. So, we were taught to write in the passive voice and without any verbs suggesting intentionality. As a result of the low-fidelity of the viral reverse-transcriptase, many mutations are accumulated in the genomic segments encoding the HIV envelope. No one is introducing mutations in the envelope to actively evade immune defences.
Although having clear scientific language can prevent us from getting confused about causality, such as actually believing that the HIV envelope uses last-second unfolding to “surprise” neutralizing antibodies, I believe that allowing ourselves to anthropomorphisme can be valuable for creative understanding of biological processes.
The standard formulation is that “a virus enters a cells and hijacks its cellular machinery”. This sentence carries inklings of intentionality, but importantly has the connotation that the virus is a nefarious agent which diverts cellular resources for its own benefit. A virus infection triggers a shift in the genetic and metabolic flux of the target cell. Quite neutral to intention. This formulation also suggests that a novel hybrid entity, corresponding to the sum of the virus and a cell with altered behaviour, is transiently induced during infection, an idea developed by Forterre with the virocell concept. Now, think of a virus usurping cellular processes or thinking that a virus could be performing immune subversion or immune irreverence. Key insights could trickle from these formulations.
An irreverent virus surpasses the boundaries of politeness in the cell that it infects- how might this manifest?
Can we quantify the extent to which the cell is burdened by tending to viral activities at the expense of its own ?
There is also growing interest in multicellular behaviour, in communities of bacteria like bio-films for example. Are the bacteria co-operating? There are clearly some fitness functions acting on levels beyond individual cells and co-ordinated phenotypic switches occurring across the population. In fact, this whole idea of evolution and biological function acting on multicellular assemblies is the key idea I created my podcast (Holobiont) to explore. The bacteria are probably not co-operating. Although by imprinting this agency on them, we can use terminology and conceptual frameworks from economics and game theory, like public goods, cheating and the tragedy of the commons. So no, coral reefs are probably not generous for providing shelter for bacteria and Listeria monocytogenes isn’t trying to betray macrophages.
Although, I don’t think it would hurt quite as much as we tell ourselves to let ourselves believe that they do!
TechBio Shower thoughts :)
