Reading List August 2020
Published:
Some more ion channel stuff, a brief piece of science-inspired music, and thoughts on the airing anime season.
Rao et al. from Oxford have been working on a neat tool that uses molecular dynamics to identify the conductance state of various ion channel pores. The tool, Channel Annotation Program (CHAP) is freely available and runs on Gromacs-2016 and some other libraries. Its competitor, HOLE has been around since the 90s and is the most widely-used program to characterize the width along a channel pore. However, the conductance state of an ion channel is not strictly determined by how small or large a channel pore is. Instead, there is a growing body of evidence that suggests that channels can be non-conductive even if their diameter is as large, or larger, than the diameter of the (hydrated) ions they conduct.
One explanation is that some of these surprising observations are due to hydrophobic gating. That is, as hydration (wetting) of the pore is a large (perhaps the largest) energy barrier to ion conduction, the loss of water in the pore (dewetting) can block conduction. Thus, wetting-dewetting transitions can gate ion channels just as narrowing or widening of the pore may sterically occlude ion movement. The inner pores of many channels contain hydrophobic residues that can simultaneously promote both dehydration and tight packing of the pore - both of these would understandably yield a closed pore.
Rao et al. systematically characterized the extent of hydrophobic gating in over 200 channel structures by quantifying their hydrophobicity, free energy of hydration, diameter, and other relevant aspects of pore dynamics and structure. By fitting a SVM (Support Vector Machine) to their results, they developed a heuristic that seems to help classify whether or not hydrophobic gating is expected in a given channel or not. Therefore, CHAP not only reproduces the main output of HOLE (pore size), but gives this additional information that HOLE does not. With cryoEM now widespread and rapidly evolving, I can easily see CHAP become a next-generation tool to replace HOLE in ion channel research.
A note on CHAP installation on Windows using Windows Bash, though; I struggled to install CHAP for many months. The only piece of advice I have to make it work is to uninstall/reinstall Gromacs-2016 and CHAP while following the installation instructions on their respective websites, until everything works. I don’t know why, but in previous attempts, I would get a slew of different errors - dysfunctional symlinks, files not found, etc. I’m by no means a programmer, and definitely not experienced in Bash or C++, so the experience was torture.
A cover of a cover for the August 2020 issue of Neuron Stefanini et al. put out a great article about how different types of hippocampal cells encode locomotive control in the mouse, and designed a cover for the journal in the style of a music score. After sharing it with the BBP (my graduate program cohort), we had a small debate about how it would sound if played, as the cover didn’t have an explicit tempo or time signature. I decided to just improvise it on the piano, and my cover of the cover was noticed by the authors. How cool is that! Science and music bringing people together. Love and peace.
Andrew Glazer and colleagues have developed a comprehensive database on clinically relevant mutations in SCN5A. I took a look at it, but unfortunately the patch clamp data are just scrapes of summary statistics (e.g. V1/2, maximal current), not the actual data files. Nonetheless, the database is really cool and definitely a great and important contribution. Moreover, it looks like their team is coming out with browsers for other channels soon, so we should be on the lookout for those as well.
- PlayMolecule and CHARMM-GUI PlayMolecule is a newer molecular dynamics-based platform targeted towards accelerating pharmaceutical research and development for protein targets, and hosts an array of fully online services ranging from protein protonation to binding pocket prediction and ligand optimization. The tools are incredibly intuitive, but unfortunately aren’t very transparent with their issues - if something goes wrong, the reason isn’t always clear.
On the other hand, CHARMM is an OG of the molecular dynamics field. The GUI was developed over a decade ago to function via the web, so no fancy GPUs are needed. There are a lot of services available, each with many options for customization. Unlike PlayMolecule, however, the interface is primarily developed to produce input files for simulation, rather than doing the simulations outright. So, you will still need to learn MD.
- A review of the Summer 2020 anime.
I’m currently watching SAO Alicization S2 and Futekigousha. Both are quite entertaining, even though the former’s storyline was a bit confusing (they introduced two random characters with zero backstory in one of the recent episodes). However, the production quality of SAO is so high, I’m willing to stay just for that. I was pleasantly surprised by the quality of Futekigousha, given that I enjoyed the novel and was afraid the anime would end up like Maou-sama Retry. There are some choppy parts, of course, but the overall animation and designs are respectable, and certainly watchable. Besides these two series, I’m saving Kanojo Okarishimasu for a binge session - I’m cringing a bit just thinking about the story, haha. The manga of this series was very entertaining, although irritating at times, so I’m really looking forward to watching the anime. I’m also saving Deca Dence (seems to be quite well-received), Re:zero S3, and Oregairu S3 - the latter two for when I watch/rewatch their respective earlier seasons. Overall, a pretty good season.