Tag: computing

Pixy computer vision for tracking mouse behavior

Pixy is an open source computer vision system. Mostafa Nashaat, Robert Sachdev, and colleagues including Matthew Larkum have developed software for use with the Pixy, that can be used to track mouse behavior, including free movement around an enclosure (top image), or track the movement of individual whiskers (bottom image), all at 50 Hz. Here’s



an hdf5/matlab file browser

Michael Graupner has coded a nice program for browsing and managing hdf5 files (closely related to MATLAB files) called hdf5Manager. And it’s open source.



RStudio – MATLAB-like IDE for R

Carson Chow recently shared his post-MATLAB suite of programming tools, and it involves RStudio, which is open-source. “I had planned to replace Matlab with Python, Julia, and R but I have found that R and Julia have been sufficient for my requirements.” It’s a nice environment and something to try if you’re looking for an



Remote, web-based analysis

Jeremy Freeman and his lab are developing tools for analyzing data using a workflow that is fundamentally more scalable not only in terms of computing power and data set size, but also in collaboration and sharing. The one-person, one-machine approach to data analysis can be highly efficient, but collaboration and scaling up computational power can



SIMA update – for 2p calcium imaging

SIMA has recently been updated (here’s the original Labrigger post): From the SIMA team: We have recently released updated versions of the SIMA & ROI Buddy tools for analysis of calcium imaging data (motion correction, segmentation, registration of ROIs across different imaging sessions, signal extraction). These new versions now support 3D datasets, allowing for analysis



Python ephys and calcium imaging analysis code

Andrew Giessel wrote some analysis code in Python when he worked in the Datta lab. He has since moved on to another venture, but he open-sourced the code. There are import routines for data from ScanImage and Ephus, but the majority of the code is acquisition platform agnostic. It’s called d_code.



More mini computers

Raspberry Pi is the most popular mini computer right now, but there are other options. SolidRun sells two. The first is a sleek cube called the CuBox-i (pictured above). The second is a barebones board, the Hummingboard.



A clock for the Raspberry Pi

The Raspberry Pi is a handy tool, but it has some quirks. One quirk is that there is no internal clock with a battery backup (something most real PCs have). Instead, it syncs with online clocks via TCP/IP. If you don’t want to rely on that mechanism, you could you use this, but then it



OSX features I wish Windows had

Migrating back to Windows after years of using OSX, there are some features I miss. Default Folder X (link) Right click on the file name in the title bar to find out what folder that file came from. Rename a file that is already open. Click and drag events around on the native calendar app



Software Carpentry teaching programming and data management to scientists

Software Carpentry teaches scientists how to program and use open source tools to manage their data and make their life easier. They’re focussed specifically on software tools, rather than types of analysis, and so what they teach is pretty general. They started in 1998 and are currently part of the Mozilla Foundation. They hold bootcamps,



Gobs of DIY TB storage

These units sell for about $5k, and hold 45 drives. Since 4 TB drives are pretty common these days and cost about $120 each, this scheme can provide about 180 TB for a grand total of (5000 + (45*120)) $10.4k. It’s an open design and 45drives is a division of Protocase, which Labrigger has mentioned



AutoHotKey for repetitive tasks

This the favorite tip Labrigger has received this week. Here it is, unedited, and in its entirety: I often use autohotkey for repetitive file naming and key remapping when I am forced to use clunky software on a windows computer I don’t plan on living on for long. It makes working with these programs much



ACQ4: A Python-based open source system for neurophysiology

Luke Campagnola, Megan Kratz, and Paul Manis recently published their in-house software for neurophysiology experiments. It’s an extensive set of tools, including multiphoton imaging, photostimulation mapping, image mosaic construction, electrophysiology, and more. Website: acq4.org



Is it really worth it to automate it?

We’ve been over this before. Get the Computation Done.



CRISPR resources from the Zhang Lab at MIT

Few things in biology take off like CRISPR genome engineering technology has recently. Feng Zhang’s lab at MIT has a design tool for CRISPR that has also rapidly matured. It helps to avoid off target effects.