Showcasing Digitised Molecular Neurobiology

GIF references: Screenshots by H Muzart, depicting content manipulated by H Muzart. Using software platforms and open-source data: VMD, PyMol, AllenInstitute.org Cell Science, NEURON ModelDB, etc.

From molecules to neural circuits to minds

A note on the concept of 'from molecules to neural circuits to minds'

     First of all, much of neuroscience is trying to relate neurotransmitters like dopamine, GABA, glutamine, etc, to macromolecular structures like NMDA and AMPA receptors. My time at UCL studying cellular biochemistry (as part of my BSc 2013-2016) helped me understand how these mechanisms underlie learning, plasticity, memory, sensation, pain, etc. I was also interested in neuroendocrinology (effects of oxytocin, serotonin, testosterone, etc) and neuropharmacological agents; as well as neurology, immunology and disease/health states. I also visited various industry and academic labs which sought to translate this animal molecular neurobiology to scaled-up neural circuits. Ultimately, we can get to an interactome at both of these levels.

     This relates to other implementations of dopaminergic neurons in artificial neural networks (in SPM DEM models, SimBrain NNs, Emergent NNs, Emergent NN dopamine, A.I. dopamine [ link-1, link-2 , link-3 , link-4 ], also as reviewed by me). Another interest of mine, for the future, and subject to budgeting and legal approvals, may also use intranasal/oral dopamine-equivalents (ordered from companies), to see the effects on humans, which I could use for participants while conducting online human experiments (for example with experiments shown here: Paradigm visual emotional stimuli decisions, Eye movements, etc) but also to see the effect on the brain (in order to gather new neural data for these: Neuroimaging, BMIs, etc) 

-----------------

In brief:

Gene(s) --> Protein(s) --> Neurons --> Neural Functions --> Cognition

Gene <--> Genome

Genome <--> Proteome

But, how can we map these levels of analysis?

genetics ---> neurology

genetics ---> polygenic psychological/psychiatric traits

[ genetics + cellular physiological bioinformatics ] ↔  neuroimaging (EEG, fMRI, etc) of brain circuits ↔  psychology

Some other projects (IMAGEN meta-studies, Crick Inst groups, SWC/Gatsby optogenetics investigations, neuroConstruct (UCL worm neuronal connectome), NeuroML language, etc) are addressing these issues

-----------------

Also see BioNeuroTech.com/bcm.

Genomic Neuroinformatics

      There are huge amounts of data out there, as evolution by natural selection over billions of years has given us something wonderful. There are huge amounts of data -- and not only have we (as humans) not discovered everything, but even with the data we do have, it is very hard to make sense of it all and analyse it. I have looked into some few of the hundreds of biotech start-ups (some being spin-offs of university labs) for phylogenetic ancestry tracing, probabilistic disease analysis, biosample orders, etc (Allen Institute, 23andMe [my profile], New England Biolabs, alt HGP) that have been set up in recent years, and many provide public-domain data (see AllenCellDB, UK BioBank, etc) that I could analyse with open-source packages (like Haploview (for GWAS SNPs), R , MS Excel, etc) which I used in my own time (during 2013-2015 at UCL and 2018-2019 at KCL) (also see my linked shared Private Cloud Drive folders); and it still remains insightful to explore the data. So far, my data manipulating remains basic for my own learning and not to advance scientific research per se, as it is not my field of expertise.

-----------------

Also see BioNeuroTech.com/bcm.

Neuronal Biophysics

     Some useful tools I used in 2017-2018 include NEURON (using datacode from senselab.med.yale.edu/ModelDB/) to help simulate axon and synapse compartments. While these basic computational models have been around for well over 20 years, I have not used these to a full extent, and will be looking into it more. Nowadays, there are even greater advances in software. Again, these are another step forward.

-----------------

Also see BioNeuroTech.com/bcm.

Computational Biochemistry

     Since Stanford's M Levitt's 2013 Nobel, I have developed greater interest in this field. However, very little of my time in UCL biochemistry labs and workshops was dedicated on computational modelling. Some useful tools I used in 2016-2018 include PyMol, VMD (UoI), etc, which help simulate macromolecules. While these models have been around for many years, I have not used these to a full extent, and will be looking into it more. Now, there are even greater advances (e.g. AlphaFold code, etc). Again, these are another step forward. 

In essence, there are 2 main points here: (1) visually-rendered representations of molecules and molecular processes help scientists understand and explore these by visually conceptualising those mechanisms at that low structural level; (2) those functional models can be experimented with for further scientific research.

-----------------

Also see BioNeuroTech.com/bcm.

Further works...

Furthering DIY home-based democratised biochemtech

     Ultimately, I am trying to help pioneer the way for the mindset of do-it-yourself home-based libertarian-minded democratised neurosciences/biosciences (including bionanotechnology biohacking), and 3D computational models of these. From my practical experiences in biochemtech in 2008-2010 and 2013-2015, I have been pursuing an interest in this. Since 2017, people have been experimenting with sub-integumentary (under-the-skin) bio-nano-implants tags, genetic bioengineering, CRISPR, mRNA, PCR, micro blottings, new centrifugation techniques, self-admin biosample tests, biolabs, chemistry of drug dev and testing, developing iPSC/stem cell growths, synthetic biotissues, and organoids; as well as petri-dish-based (synthetic biological artificial neurons) neuronal ensembles and neuromorphic computations; these, since 2020, can now be set-up with relatively cheap raw materials in the bedroom/garage/etc. Access to data and the right physical tools and resources is to be democratised. This however, could provide huge threats, e.g. pathogenic de novo viruses, biohacking, etc. My further interest would be to contribute to computational models of all the above; as well as ethical infrastructures.

More info will be posted here in the future.

External links: TBC.