The molpigs Podcast

Today we are joined by Lee Organick, a PhD student in the Molecular Information Systems Lab (MISL) at the University of Washington. Lee is a biologist turned computer scientist and engineer, quite a unique transition! She explains how she was “forced” to take a computer science class in her undergrad, which opened up a completely new field of interests. After this, she started incorporating more programming into her research, and as such slowly moved into more computational fields. This is how she eventually found herself at MISL, and has been programming molecules ever since. She talks about how the transition from biology to computer science was a difficult one, and how she suspects that she invested more time than the average student moving in the opposite direction.

We then move on to talk about her main research area, DNA storage. Because we focussed on the specifics of DNA storage in our previous episode with Yuan-Jyue Chen (another member of MISL), we spoke more about the future of DNA storage, specifically where it fits in to the current data storage ecosystem. Lee argues that DNA is currently best suited archival and long term data storage, with many advantages over the traditionally used tape. Lee also talks us through a new project she is working on which involves augmenting the previously discussed DNA based image similarity search with Cas9!

Finally we move onto an extremely interesting topic; that of the security concerns surrounding DNA data storage. We learn about how current sequencing machines may be vulnerable to buffer overflow attacks through the submission of malicious sequences, how sequencing leak can result in a nefarious third party being able to “spy” on other people’s pooled sequences, and some of our committee members even suggest some new potential exploits!

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/organick-l-48ebaeff2751376d/

Join us this week for an extremely interesting conversation with Kent Kemmish, the founder and chief exorcist (yes, exorcist) officer of Molecular Reality, and the creator of the new, and world’s first “molecular” games console, the demonpore 64. This is our first podcast with a member of the molecular programming community who works in industry, and in the startup sector.

At its heart, Kent’s demonpore 64 is a device which utilises solid state nanopores in order to sense its environment at the molecular level. Kent and his team have reduced the cost to manufacture this traditionally lab based and expensive equipment by orders of magnitude.

Kent’s hope is that by making this technology cheap, accessible and gamified, gamers can engage in science and data collection on behalf of scientists. This ought to enable scientists to perform very very large experiments, which would otherwise be impossible. We discuss what a molecular games console actually is, what sort of functionality the demonpore 64 has and how it can bring together academics, game developers, gamers, and citizen scientists in order to solve the world’s biggest challenges.

Kent talks through future games which will be available for the demonpore 64, including Poop of the Gods, 2021: A SARS Odyssey, Genomic Ranger, and Molecules of Mars. These games allow people to explore the molecular mysteries of poop, viral detection, chromosome structure, and martian soil!

We then move on to talk about Kent’s life, from student, to academic, to entrepreneur. Kent talks fondly of his early days working on Drosophila genetics, and later at Halcyon Molecular a startup which was focussed around sequencing DNA with electron microscopy. He talks about how these experiences shaped his worldview around having a much greater impact by working on and caring about scientific tools rather than the science itself, and bringing these tools to the scientists with research questions.

To learn more about the demonpore 64, check out Kent’s WeFunder campaign https://wefunder.com/molecularreality, and also check out their website https://www.molecularreality.com/.

The nanopore cartridge shown at ~51:00 can be seen here: https://podcast.molpi.gs/media/kemmish-k-03808e4f7ec87ae1/nanopore-cartridge.jpg

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/kemmish-k-03808e4f7ec87ae1/

What do ant colonies have to do with molecular programming? In this podcast, we spoke with Namita Sarraf, a graduate student at Caltech in Lulu Qian’s group.

We discuss her research, which revolves around the production of multifunctional and modular DNA robots. Namita takes inspiration from ant colony dynamics to design robots, which alone may exhibit simple behaviour, but show emergent complexity when put together. By having these robots pattern the surface, ant pheromones can be emulated. One task which these “DNA ants” are being made to perform is maze-solving. Because traditional methods are not ideal for DNA robots, Namita is developing bespoke maze-solving algorithms. As she points out however, maze-solving by itself is not inherently useful, and for this reason these DNA robots are being built for modularity and composability. By combining maze-solving with cargo sorting Namita can generate more complex behaviours with real world applications.

We then move on to talk about how Namita moved into molecular programming from her original field of tissue engineering. We discuss graduate student life, impostor syndrome, and the generation of negative results and their use in publishing.

Namita is also one of the founders of the open collaborative textbook project “The Art of Molecular Programming”, a grassroots project aimed at collecting experts in the field to build a comprehensive textbook which will serve as a starting point for new and existing researchers. We discuss how the idea came about, inspired by the spirit of the Synthetic Biology community. The Art of Molecular Programming aims to be a project which collects all of the useful pieces of lore which exist scattered throughout the molecular programming literature and put them in one useful repository, taking away the pain that new graduate students endure in their first years while they struggle to build up a coherent picture of the field by reading countless ad-hoc papers.

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/sarraf-n-5f564ce4c08e6e5c/

Kate Adamala is a biochemist building synthetic cells at the University of Minnesota College of Biological Sciences. Her research aims at understanding chemical principles of biology, using artificial cells to create new tools for bioengineering, drug development, and basic research. The interests of her lab span questions from the origin and earliest evolution of life, using synthetic biology to colonize space, to the future of biotechnology and medicine.

She received a MSc in chemistry from the University of Warsaw, Poland, studying synthetic organic chemistry. In grad school, she worked with professor Pier Luigi Luisi from University Roma Tre and Jack Szostak from Harvard University. She studied RNA biophysics, small peptide catalysis and liposome dynamics, in an effort to build a chemical system capable of Darwinian evolution. Kate’s postdoctoral work in Ed Boyden’s Synthetic Neurobiology group at MIT focused on developing novel methods for multiplex control and readout of mammalian cells. Her full first name spells Katarzyna; she goes by Kate for the benefit of friends speaking less consonant-enriched languages.

First we discuss Kate’s synthetic cells and whether or not they are living. These are phospholipid liposomes which encapsulate a full central dogma (transcription, translation). Synthetic cells are more complex than biochemical experiments, but at the moment, Kate does not consider her synthetic cells living. These cells are not self replicating, currently requiring a graduate-assisted replication. We then have an extended discussion about the ribosome, why it’s the biggest hurdle to achieving true self replication, and why it kind of sucks as a catalyst!

Next, we move on to how synthetic cells can be used to aid in the research of brain computer interfaces (BCI). Kate’s vision is that, because synthetic cells can be so robustly controlled, they represent a form of “programmable goo” which would interface much more robustly with our brains than traditional silicon. She envisions the role of synthetic cells as being used as a less injurious interface for BCIs, which currently cause significant scarring to the brain.

Finally, we talk about one of the most interesting topics covered on the molpigs podcast: space exploration! Kate discusses how synthetic cells, being so programmable, might be ideal devices for Martian terraforming. By engineering poly-extremophiles (extremophiles which are robust to many extreme conditions, organisms which do not exist on Earth) specific to the environment of Mars, it may be possible to design a metabolism capable to transforming Martian soil into something fertile. Additionally, synthetic cells might be used as on-board biochemical printers on long space missions. Their programmable metabolism may enable us to produce any biomolecule, such as medicines on demand.

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/adamala-k-c1a694437dacfe8e/

In the third episode of our ‘Lab Pigs’ series, which highlights the research and journeys of early career researchers in our field, we talked with Erik Poppleton, of the Biodesign Institute at Arizona State University. Erik researches the use of computational modeling in informing the design of molecular machines. As part of this, he also develops general-use analysis tools for oxDNA, and conversion tools to integrate the various design and simulation tools in the nucleic acid nanotechnology ecosystem. We talked about his research, his experience writing academic software, and the relationship between geology and molecular programming.

Core Simulation Tools
- Main oxDNA documentation: https://dna.physics.ox.ac.uk/index.php/Main_Page
- Current stable release (being retired soon): https://sourceforge.net/projects/oxdna/files/
- Bleeding edge release (has Python bindings!): https://github.com/lorenzo-rovigatti/oxDNA
- The model is also available as part of LAMMPS, documentation can be found here: https://lammps.sandia.gov/doc/Packages_details.html#pkg-user-cgdna

Useful tutorials
- A textbook chapter covering how to relax and simulate origamis: https://arxiv.org/pdf/2004.05052.pdf
- A textbook chapter covering the details of molecular simulation: https://www.public.asu.edu/~psulc/myimages/chapter.pdf
- Example input files: https://github.com/sulcgroup/oxdna_analysis_tools/tree/master/example_input_files

Useful tools
- TacoxDNA, converters from design software to oxDNA: http://tacoxdna.sissa.it/
- oxView, a visualizer and editor for oxDNA: https://sulcgroup.github.io/oxdna-viewer/
- oxView documentation: https://github.com/sulcgroup/oxdna-viewer
- oxdna_analysis_tools, a library of python scripts for basic simulation analysis: https://github.com/sulcgroup/oxdna_analysis_tools
- oxdna.org, a public webserver for running simulations: oxdna.org
- ox-serve, run interactive simulations in your web browser using a Google Colab GPU: https://colab.research.google.com/drive/1nFC9zy-wEwwl8vlJZAbQZZofavP4PXvL#scrollTo=C_8TB2t5gxDg

Of course, if you find these tools useful, please remember to cite us! The citations for each tool can be found in its documentation (oxdna.org paper coming soon!)

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/poppleton-e-d83cc0825a805d2e/

In this episode we talked with Yuan-Jyue Chen, of Microsoft Research and the University of Washington, on some of his research into DNA Data Storage. Yuan focussed on two topics: random access of data, and the accompanying issues with stochasticity and errors, and an application of DNA storage for efficiently searching a large database of images by similarity.

Please note: The views expressed by Yuan in this podcast do not necessarily represent the views of Microsoft.

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/chen-yj-0e6debec064e5f0d/

In the second episode of our 'Lab Pigs' series, which highlights the research and journeys of early career researchers in our field, we talked with Josie Kishi. Josie was instrumental in developing the Primer Exchange Reaction (PER) synthesis method and the related imaging method, SABER. As well as talking about these, we found out what excites her about molecular programming, how she got into the field, and where she things it's going to go.

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/kishi-j-db823daf0f863ab6/

Join us this week for a long and interesting conversation with Tom Ouldridge of Imperial College London on Maxwell’s demon, Szilard’s engine, what people get wrong about thermodynamics and information theory, how this all relates to biology, and how his lab is using these ideas to develop exciting new approaches to molecular programming.

Tom Ouldridge is a Royal Society University Research Fellow in the Bioengineering Department, where he leads the “Principles of Biomolecular Systems” group. His group probes the fundamental principles underlying complex biochemical systems through theoretical modelling, simulation and experiment. In particular, they focus on the interplay between the detailed biochemistry and the overall output of a process such as sensing, replication or self-assembly. They are inspired by natural systems, and aim to explore the possibilities of engineering artificial analogs.

We start by discussing Maxwell’s demon and Szilard’s engine—thought experiments from the 19th and early 20th centuries about the interplay of thermodynamics and information-processing. These have long captured the imagination of theoretical physicists. There is renewed interest in these thought experiments due our increasing ability to control systems at the molecular level. Many still disagree about the interpretation of these ideas, the implications for the second law of thermodynamics, and the consequences for thermodynamics of computation.

Szilard’s engine is a simpler version of Maxwell’s thought experiment, but which is mathematically tractable, considering only a single particle separated by a divider attached to a weight. If the particle and the weight are on the same side, then the particle can bounce against the divider and lift the weight, doing work. By resetting the divider, this step can be repeated to extract more work. Tom talks about how this seeming paradox may be resolved.

Tom discusses how his group has implemented a theoretical Szilard engine in biomolecules; by explicitly rendering each step of the engine as a biochemical process (using cell surface receptors). This helps demystify the whole process by rendering all “information theoretic” steps as concrete, real, processes. Doing so is helpful not only in resolving old thought experiments, but because the crucial idea—that the generation of correlation between non-interacting degrees of freedom is thermodynamically costly—is of fundamental significance to natural and synthetic molecular information-processing systems.

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/ouldridge-t-4007264116dd3097/

Join us for the first of our ‘Lab Pigs’ series, in which we talk with early career researchers on their research and journey within our field. In this episode, we chatted with Dominic Scalise. We talked a lot with Dominic about his work towards building a stored program chemical computer.

Dominic Scalise is a postdoctoral scholar in Lulu Qian’s lab at Caltech. He earned his PhD in chemical and biomolecular engineering from Johns Hopkins, advised by Rebecca Schulman, and his B.S. in mechanical engineering from UC Berkeley. His work focuses on developing a stored program chemical computer, and powering circuits using DNA buffer reactions.

Abstract: Molecular programming extends computer science beyond electronics into chemistry, and lets humans directly program physical matter. However, chemical circuits remain arduous to program, often requiring months or years to design, implement, and debug even a single program. In stark contrast, we program modern electronics with much less effort. A critical step in simplifying electronic programming was the invention of “software” in the 1940s. I will outline how similar concepts of “chemical software”, in which programs are stored in memory rather than hard coded into the connections of chemical reaction networks, could dramatically simplify the task of chemical programming. I will then discuss some reaction motifs in development which may be useful for implementing chemical software.

In this podcast, we found out that current chemical computers are just like the electronic computers from the 1940s, in that a computer program requires rewiring the hardware. They are small in size, error-prone, and take a long time to build. A stored program chemical computer may tackle these problems by being a robust universal circuit capable of running arbitrary algorithms, with the exact algorithm of a given computation depending on which instructions (software) are given.

Dominic then discussed his approach—building a chemical memory and then a chemical processor, possible challenges in doing so, and his vision for the future. We also talked a bit about his experiences in academia.

Dominic made a special announcement for molpigs members at the end about a grassroots project: a Molecular Programming textbook to drive the field through a collaborative approach. Stay tuned on our newsletter for more information!

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/scalise-d-5cbae63946bde0c4/

Following on from Brenda's fantastic tutorial, we chatted with her to get answers to many questions, find out more about her lab's work, and get her thoughts on the future direction of this approach!

---
Find more information at the episode page here:
https://podcast.molpi.gs/media/rubenstein-b-2b86de754345/