Did you miss the Society for Neuroscience 2019 meeting? Check out our Editor Francesca Lake’s highlights of the meeting.
This year’s Society for Neuroscience (SfN) conference in Chicago (IL, USA) continued with last year’s trend, with a large focus on improving analysis of the huge amount of data we are now able to generate. This was evidenced in a number of talks and posters, as well as by many of the exhibitors showing at the event. Reproducibility also remained a key theme. However, the research agenda remained broad overall. 27,832 people attended from 75 countries, with over 13,000 abstracts presented and over 800 sessions. It’s impossible to cover the entirety of such a large meeting, but here are my highlights from this year.
Topics and Presentations
Is an artificial brain key to understanding your own?
‘Artificial Intelligence’ is a phrase we heard a lot at SfN 2019 and indeed mention a lot within the pages of BioTechniques. One of our highlight sessions was “Artificial Intelligence and Neuroscience: From Neural Dynamics to Artificial Agents”, chaired by Maneesha Sahani (University College London, UK). This minisymposium saw an array of speakers discuss artificial intelligence and machine learning approaches to understand both structure and computations in the brain.
There was also a poster session dedicated to virtual brain models. A group of posters that particularly caught my eye described TheVirtualBrain (TVB), a ‘neuroinformatics’ platform that adds on biophysical knowledge to conventional brain imaging. Presented by the Ritter lab at UT Dallas (TX, USA), this could be very interesting for precision medicine.
In other efforts to understand our own brain functions, a series of researchers presented their work examining social and emotional behavior using neuroimaging techniques. A full round-up of those talks is available here.
Delving further into behavior, F. Javier Rubio (NIDA IRP, MD, USA), discussed his work looking to understand mechanisms underlying addictive behaviors, which is focussing in on the synapse. At SfN 2019 he presented work validating a flow cytometry approach using isolated synaptoneurosomes to examine protein changes during drug use. “We call [the method] FASS, for fluorescence-activated synaptoneurosome sorting!” He told us.
We have an interview with Rubio coming soon. Register now to receive an update when this becomes available!
Reproducibility and open science
I’ve had many conversations regarding the ability to ensure artificial intelligence is reproducible, given it might be difficult to control its learning. This isn’t specific to artificial intelligence, and reproducibility in the life sciences is a key current focus for BioTechniques. At SfN 2019, we attended an interesting workshop entitled “Reproducibility for Everyone”, a collaboration between eLife, Addgene, protocols.io and CodeOcean. The session looked at tools enabling reproducible workflows for organization, documentation, analysis and dissemination. There was a real focus on the use of electronic lab notebooks; while this seems like a simple concept, the ease of the conventional paper lab notebooks often allows researchers to fall back into their familiar territory. This part of the session aimed to save time and resources in the long term and provide researchers with the framework to make research more easily accessible, searchable and shareable. You can access the resources presented at the workshop here.
It was nice to see over 50 posters covering reproducibility, too. These posters presented new software tools (for example, Open Science Chain presented by Sivagnanam et al. from UCSD [CA, USA]) to enable better reproducibility, and results of studies looking into the reproducibility of current techniques (for example an examination of transcranial direct current stimulation, performed by Takahashi and Yotsumoto from The University of Tokyo, Japan). This is particularly good to see considering that research ensuring reproducibility has traditionally been considered the less glamorous side of the lab and has been avoided as a result.
Not only was there a good focus on reproducibility, but also on open source tools, with a specially curated itinerary for the topic. Plenty of open computational tools and software were presented at SfN 2019, including those for data analysis, and we applaud the efforts being made in this area.
Targeting neurodegenerative diseases
Alzheimer’s and other neurodegenerative diseases were of course subjects of high interest to many of the SfN delegates, and some very interesting research was presented. Within this, microbiomics was one of the key topics up for discussion, with multiple studies being presented that investigated how gut–brain communication affects murine brain health, in particular regarding Alzheimer’s disease, as well as the implications for cognitive neuroscience.
Gene therapy advances were also highly discussed, with researchers presenting promising results from mouse models that used adeno-associated virus-based gene therapy to treat Amyotrophic Lateral Sclerosis, Alzheimer’s disease and Parkinson’s.
It’s unsurprising CRISPR was a subject of discussion at SfN 2019, not least due to the meeting’s proximity to the announcement of prime-editing. The CRISPR-based research presented at SfN and pertinent to neurodegenerative diseases was fascinating. The Kampmann lab (University of California,San Francisco; CA, USA), for example, has developed a new genome-wide CRISPRi and CRISPRa screening platform for human iPSC-derived neurons. Kampmann presented the first application, uncovering genes that modulate tau aggregations. “We’re very excited to extend this platform to other cell types relevant for brain function and disease including iPSC-derived astrocytes and microglia, and to then combine these different cell types in organoids,” Kampmann explained.
We have a video interview with Kampmann coming soon. Register now to receive an update when this becomes available!
Our tech highlights from this year
One of my favourite things about SfN meetings is learning about how the plethora of exciting, newly available technology work. Here’s my highlights from this year!
Imaging the brain
Aiforia (Helsinki, Finland) announced an exciting partnership with the Neuroscience Associates (NSA) where they’re combining advanced neuro-histology and artificial intelligence image analysis in a bid to create a whole-brain reference dataset for neurodegenerative diseases. I spoke with Thomas Westerling (Aiforia), and it seems a very exciting and complicated project.
“The last year has had a lot of disappointments when it came to progress in neurodegenerative disease, both in terms of research projects and therapeutics,” Westerling told me. “One of the reasons for this is the lack of a comprehensive atlas and analysis of larger, human specimens. NSA has this wonderful capacity and talent to section and process whole-hemisphere pieces of the brain.” He noted that the project is in its early days, and very ambitious. However, the combination of NSA’s skills and Aiforia’s strengths are promising. “We are powered by state-of-the-art AI, or deep learning, architecture where users can deploy and train almost human-like pattern recognition expert systems. The flexible and robust infrastructure can handle gigapixel images and the data load, and avoid classical problems like over-training. These two technologies together can generate something we have not seen in neuro research before.”
Aiforia has deployed over 200 AI models for various projects, and it can be used when looking for anything that a human being can recognize. Westerling is excited about the benefits the AI can bring in terms of human labor cost: “the researchers themselves that have previously done this tedious manual labor can go on and do things that humans are uniquely capable of.”
- Find out more about this project in our upcoming video with Aiforia. Register now to receive an update when this becomes available!
Zeiss (Oberkochen, Germany) were also at the meeting and showed us some of their latest tech that is designed to increase automation and reduce time commitments. Their Smart Microscopy and Celldiscoverer 7 systems were particularly interesting, developed with a view to reduce time needed for training, and improve the reproducibility of results through increased automation. We also enjoyed how the Smart Microscopy system can easily transfer images to a tablet, making it more useful in educational situations.
Inscopix (CA, USA) had also announced some exciting new developments and partnerships enabling the use of its mini microscope and brain mapping platform in freely behaving models of neurodegenerative diseases. At SfN, they presented the first successful use of the technology in non-human primates. I spoke to Founder and CEO of Inscopix, Kunal Ghosh and Senior Lead Scientist, Jonathan Nassi about this as well as their newly announced partnerships. One is with the Broad Institute (MA, USA), combining the platform with Broad’s expertise in single-cell transcriptomics, with the aim of developing a better understanding of Parkinson’s disease and identify new targets for therapeutic development. The other is with Astellas (Tokyo, Japan), which is looking for new drug targets in psychiatric disorders. In addition to broadening the use to non-human primates, they’re also working on innovations in the process and data analysis to make it more streamlined, interpretable and reproducible. “Neuroscience has already moved into this generation where these kinds of circuit-level neuroscience experiments are very multidisciplinary,” commented Nassi. “Some of the big labs have all the pieces to go at it – and the expertise – but we’re reaching well beyond those labs, to a lab that is maybe just doing behavior in rodents, and has been watching this and going ‘wow I would love to do this’ but it’s never been possible and there’s been too many barriers. We’re trying to take down those barriers to entry and expand the community that can do this kind of parallel science.”
Read more about the Inscopix platform and the exciting new research projects in our interview with Nassi and Ghosh. Register today to receive a notification when this goes live
New tech for simplicity and reproducibility
Lonza (Basel, Switzerland) were also showcasing their Nucleofector tech for transfection. While the tech is well established and was used by the Church lab to develop CRISPR-Cas9, it also has interesting applications in neuroscience. Bill Busby (Lonza), noted that researchers are “using this device to transfect primary neurons without the need for excessive cell death or use of a viral method.”
Nidhi Vashistha from Bio-Techne showed us their exciting new tech, the RNAscope™, which seeks to address the unique difficulties posed by analyzing gene expression within the complex tissues of the nervous system. Nidhi was particularly pleased with the reaction they’d had from researchers, who were excited by the ability to get all the data required from a single sample, reducing the impact of sample-to-sample variation on results.
We have a video on the RNAscope coming soon, but in the meantime check out this short interview https://twitter.com/MyBioTechniques/status/1187091785757216769.
By-far my favorite technology presented at the meeting, however, was from Backyard Brains, which was showcasing tech they use to inspire the future generation of neuroscientists. Take a look at my colleague Sharon Salt controlling a very uncomfortable Joseph Martin’s muscle movements in this video.
Ever wanted to physically be in control of someone? Watch our Editor @sharonsalt take control of Joe Martin from @MyBioTechniques here at @BackyardBrains at #SfN19 #SfN2019 @Neurosci2019 pic.twitter.com/tRSaudxhXR
— Neuro Central (@Neuro_Central) October 23, 2019
SfN will be celebrating its 50th meeting in 2020, which is taking place in Washington DC, October 24—28. The BioTechniques team will be there and covering the meeting, so make sure you check back with us for more highlights in 2020!