Janelle Pakan
Carsen Stringer (she/her)Make your next discovery using #Rastermap, a visualization method for large-scale neural data in #python. Paper now out: https://www.biorxiv.org/content/10.1101/2023.07.25.550571v1 (CLICK ON THE GIF) #neuroscience
You can explore your data in the graphical user interface: #Rastermap
#Rastermap finds single-trial sequences of neural activity in a virtual reality experiment:
#Rastermap finds movement-related structure in spontaneous activity in complete darkness:
#Rastermap sorting of #hippocampus data from Grosmark & Buzsaki, 2016:
#Rastermap sorting of wholebrain #zebrafish activity from Chen et al, 2018:
#Rastermap sorting of neurons from #ReinforcementLearning agents playing Atari games:
Learn more on our github: https://github.com/MouseLand/rastermap. #Rastermap is fast thanks to #python #numpy, #scipy, #numba, and #scikit-learn. The GUI is powered by #pyqt and #pyqtgraph, and supports npz, npy, mat and #nwb ophys files.
Excited to see new datasets explored with this! If you have issues, please post an issue on the #Rastermap github: https://github.com/MouseLand/rastermap/issues.
kevinbolding@computingnature Here are some OB glomeruli rastermapped while the mouse is running around smelling things in the little colored circles. Definitely helped convert a noisy mess into something that makes sense.
manishawow, this happened within a day of announcing this #OpenSource project! thank you for sharing this @kevinbolding
Congratulations and thank you @computingnature for sharing both the source code for #Rastermap and previous #OpenData
So stoked and look forward to this becoming the norm where scientists don't have to keep reinventing the wheel in their own labs
#neuroscience #AcademicChatter @academicchatter @academicsunite
@manisha @computingnature @academicchatter @academicsunite ah well this didn’t happen in a day. there was some example Matlab code that had been made available previously. glad it’s gotten a more formal grounding/distribution framework though now.
El Duvelle@computingnature this looks super cool and useful!! Congrats and thank you!
jonny (good kind)@computingnature
this looks really nice and I look forward to giving the code a read :) might be fun to do code review with @pyOpenSci (in addition to wherever else yr sending it), really lovely process and I would definitely volunteer to review to have a reason to give it a close read ❤️❤️
this looks really nice and I look forward to giving the code a read :) might be fun to do code review with @pyOpenSci (in addition to wherever else yr sending it), really lovely process and I would definitely volunteer to review to have a reason to give it a close read ❤️❤️
@jonny @pyOpenSci are they interested in packages that are also undergoing the conventional journal review process? It seems like a great initiative for packages that aren't undergoing that process and in general to get code feedback
@computingnature
@pyOpenSci
I am pretty positive yes - it's much more about code review than I have seen be typical of journal review of code papers. huge variation there ofc, I've done reviews for journals that only cared about validating the claims made in the paper re: results and didn't care about code review at all, but others did.
so ime it is complementary! and part of it is also about helping bridge best practices from those ppl very involved with core python ecosystem with researcher-programmers to make a curated index of packages that meet some minimal standards for usability, documentation, and testing.
just an idea! obvs it's extra work but it's way more fun work than trad peer review :) (and it's a really lovely group of people to yno talk about ur work with)
pyOpenSci@jonny @computingnature hey Carsen! it is ok if your package is under Journal review. we are not a publisher, we focus on code and packages that will be useful to the open science community over time. The main scope items are:
1. it needs to be a package that you do plan to maintain after review.
2. it needs to be in our scope which you can read more about here - https://www.pyopensci.org/software-peer-review/about/package-scope.html
pls feel free to ask any other questions!!