open-source electrophysiology

Filtering by Author: Josh Siegle

Share

October 2019 Newsletter

Added on by Josh Siegle.

Open Ephys at SfN 2019


It's going to be a busy week for us at this year's conference! We're organizing our first-ever booth (in collaboration with the UCLA Miniscope team), as well as putting together a number of posters, satellite events, and a DJ set. Read on to find out all the details...

3d05b9d9-67ca-4b14-ae46-5fd806d7bc3c.png

The best place to find us this year will be at the Open Ephys x Miniscope Booth (#1766). We'll be showcasing oni, or "open neural interface," a new data acquisition system that will work seamlessly with our next-generation tetrode headstages, Neuropixels, and the latest Miniscopes. This will be an incredibly versatile tool, no matter where your science takes you. It comes in PCIe and USB versions, and has a whole host of features you'll want to explore. If you won't be at the meeting, be sure to check out our "virtual booth" by following this link.

Open-Source Technology Social

Jakob Voigts and Denise Cai are organizing a social for people using open-source tools and those curious to learn more about them. Come say hello on Monday evening from 6:45-8:45 pm in room N135 of McCormick Place.

Tools & Tech: A BRAIN Initiative Alliance Social

Open Ephys will have a table at this event, along with numerous other toolmakers in the field. It's happening Sunday evening from 6:30 - 8:30 pm at the Hyatt Regency McCormick Place Ballroom CD. More info can be found on the event website.

Poster sessions

Our main focus this year is on our booth, but there are also a few key posters that highlight the open-source tools we're using and developing:

Gonçalo Lopes - “BonVision - An open-source software to create and control visual environments” (Monday morning, L14)

Pavel Kulik – “Real-time processing and visualization of high-channel-count electrophysiology data with the Open Ephys GUI” (Tuesday morning, CC52)

Aarón Cuevas López – “A low-power wireless transmission system of neural data by hardware compression” (Tuesday morning, CC53)

Ephys data acquisition software survey

We're launching a new survey to help guide our software development over the next year. If you regularly carry out experiments with extracellular electrodes (regardless of whether or not you use Open Ephys tools), you are highly encouraged to respond! One randomly selected participant will receive a $100 Newegg gift card. Please visit survey.open-ephys.org to fill it out; it should only take 5-10 minutes.

"Meet the Experts" at the Neuropixels Booth

If you're currently using Neuropixels—or thinking about starting—there will be plenty of opportunities to ask questions at the Neuropixels booth (#2037). Josh Siegle will be there discussing Open Ephys/Neuropixels integration Sunday from 4-5 pm, Monday from 1-2 pm, Tuesday from 4-5 pm, and Wednesday from 9:30-10:30 am. Josh will also be at the Allen Institute booth (#401) answering questions between 10:30 and 12:30 on Sunday.

Neurolabware Party

Our friends at Neurolabware have consistently thrown THE party to attend at SfN, and this year is no different. The event is happening on Tuesday night from 9 pm - 2 am at PRYSM Nightclub. They aren't giving out wristbands this year, so be sure to arrive early! In any case, you'll want to get there to catch Josh Siegle's DJ set, which starts right at 9.

Compact, Open Ephys compatible headstages

At their booth this year (#1669), Cambridge NeuroTech will be showcasing *very tiny* 64-channel headstages that are plug-and-play compatible with the Open Ephys acquisition board. We are huge proponents of open interfaces, and we're glad that Cambridge Neuro decided to adopt Intan's SPI standard for these, which only weigh 0.38 g.

Open Ephys GUI updates

There's a new version of the Open Ephys GUI available for download (version 0.4.5). The major updates include improved support for Neuropixels (Windows only) and migrating the build system to CMake, which will allow us to use continuous integration for application testing and distribution. A big change on the developer side is that all plugins that rely on external libraries (e.g. ZeroMQ, HDF5) have been moved to separate repositories, so you can build the GUI with Juce as the only dependency. If you have any questions about the changes, don't hesitate to write to info@open-ephys.org!


Until next time,
The Open Ephys Team

Share

November 2018 Newsletter

Added on by Josh Siegle.

Open Ephys at SfN 2018

For those of you going to San Diego this year, here are the opportunities to hear more about what we've been up to lately:

At the Monday afternoon session, Jon Newman will be presenting a poster on a PCIe-based platform for real-time feedback experiments. The specs on this thing are pretty incredible—it's easily the most advanced system for high-density tetrode recordings in freely moving rodents. Furthermore, the design includes a general-purpose API for low-latency data transmission, which will facilitate closed-loop experiments with all types of hardware.

We'd love to hear your thoughts on how this system should evolve. Please fill out this survey, and come talk to us on Monday!

431.24 / LLL59: Open Ephys++: High performance open-source firmware, APIs, and hardware for closed-loop neuroscience experiments
November 5
1 - 5 pm
Poster hall
https://abstractsonline.com/pp8/#!/4649/presentation/24889

On Tuesday evening, we'll be at the Open-Source Technology Social along with a host of other people developing and disseminating open tools for neuroscience.

SS29: Open-Source Technology Social
November 6
6:45 - 8:45 pm
Marriott Marquis - Grand Ballroom 10
https://abstractsonline.com/pp8/#!/4649/session/272

We also recommend checking out this list of posters and presentations about open-source tools compiled by the folks from Open Behavior.

Neuropixels: 384-channel silicon probes for multi-area recordings

This week saw the official public release of Neuropixels, a new type of silicon probe that is poised to have a major impact on the field of electrophysiology. These probes, which were developed by imec in collaboration with scientists at the Allen Institute, HHMI Janelia Research Campus, and UCL, contain more recording sites per shank than any other implantable neural recording device. Their high channel count makes it possible to record from many structures along the length of the probe, while their compact size makes it easy to insert multiple Neuropixels into the brain at once. They have already facilitated experiments that would have seemed like pipe dreams just a few years ago. For example, the Allen Institute is using Neuropixels to simultaneously record spiking activity from 9 nodes in the mouse visual system, including V1, LGN, superior colliculus, pulvinar, and 5 higher visual areas.

We expect that many members of the Open Ephys community will want to use Neuropixels, so we've tried to make it as easy as possible to integrate them into existing workflows. The Open Ephys GUI can stream data from any type of Neuropixels hardware, so you can take advantage of existing GUI plugins or develop your own.

To obtain probes, you’ll have to fill out the form available at neuropixels.org. There’s a minimum order size of 20 probes in 2018, with the first 30 orders filled by lottery. In 2019, imec expects to ship 2,400 probes, with the minimum order quantity reduced to five.

If you'll be at SfN, you can learn more about Neuropixels at the Allen Institute booth or check out one of over a dozen posters that include Neuropixels data.

Until next time,
The Open Ephys Team

Share

September 2018 Newsletter

Added on by Josh Siegle.

New product: pyControl

We are pleased to announce that a new open-source device is now available to be purchased through the Open Ephys Store. pyControl is a software/hardware package for implementing rodent behavior experiments, based on the Micropython microcontroller. The USB-based pyControl interface can be hooked up to nose pokes, LEDs, solenoids, audio drivers, and other devices used for behavioral training. A single computer can control many pyControl setups, allowing high-throughput experiments to be implemented at low cost.

From the pyControl documentation:

“pyControl makes it easy to program complex behavioural tasks using a clean, intuitive, and flexible syntax for specifying tasks as state machines. User created task definition files, written in Python, run directly on the microcontroller, supported by pyControl framework code. This gives users the power and simplicity of Python for specifying task behaviour, while allowing advanced users low-level access to the microcontroller hardware.”

pyControl is in daily use in a number of laboratories and has run many thousands of hours of behavior experiments. It's also a great tool for courses, given how easy it is to get started using it.

The system is being actively developed by researchers at the Champalimaud Foundation and Oxford University. Profits from the sale of pyControl will support the continued development of the system. For any questions please contact the pyControl Google Group.

New version of the GUI

Version 0.4.4 of the Open Ephys GUI is now available for download from our website and GitHub. This version includes a number of bug fixes and user interface improvements, including:

  • A new plugin for streaming data from Intan's 1024-channel Recording Controller

  • An improved Pulse Pal plugin that allows the user to configure the device from software

  • An updated FrontPanel DLL that improves performance of our acquisition board on Linux machines

We recommend that all users upgrade to this version. As always, be sure to carefully test any features you need before using this software in your experiments.

This version has been improved by collaborative development more than any previous release. We’d like to thank Alessio Buccino, Ethan Blackwood, Iryna Yavorska, Kevin Boergens, Ben Acland, Martin Spacek, Jon Newman, Evan Matteson, Ronny Eichler, Clayton Barnes, and Chuck Holmes for their invaluable help with testing and bug fixes. And of course, this release would not have been possible without the hard work of our lead support person, Aarón Cuevas López.

Preprints from the Open Ephys community

In the last 2 months, there have been at least four new bioRxiv preprints based on data collected with Open Ephys tools:

A team from the Allen Institute led by Corbett Bennett and Sam Gale has posted a preprint describing their detailed characterization of the lateral posterior thalamic nucleus of mice. They found that this nucleus contains three sub-regions, each with its own distinct anatomical connections, cellular tuning properties, and maps of visual space. All of the data was collected with “Phase 2” Neuropixels probes using the Open Ephys acquisition board.
https://www.biorxiv.org/content/early/2018/08/18/395244

Nikolaos Karalis and Anton Sirota from Ludwig-Maximilian University Munich have shown that long-range coordination between the components of the limbic system is orchestrated by respiratory rhythms. Breathing creates an oscillatory scaffold that may facilitate information transmission between disparate regions of the rodent brain. They used chronically implanted tungsten electrodes and silicon probes streaming data through the Open Ephys acquisition board.
https://www.biorxiv.org/content/early/2018/08/16/392530

Michael Okun (now at University of Leicester) and his colleagues at UCL have released a preprint characterizing the dynamics of prefrontal neurons across a wide range of timescales. They find that coordination between individual units and their neighbors can be quite distinct, depending on the timescale under investigation. Much of the data was collected with chronically implanted silicon probes using the Open Ephys acquisition board.
https://www.biorxiv.org/content/early/2018/08/18/395251

Hyeyoung Shin and Chris Moore at Brown University have posted their recent findings on gamma oscillations in somatosensory cortex. They describe a class of fast-spiking interneurons that fire at gamma frequencies but are not strongly driven by sensory inputs. The precision with which these neurons are time-locked to gamma on a given trial is correlated with detection performance in mice trained to respond to whisker deflections. The authors used flexDrives and Open Ephys acquisition boards to collect all the data in this study.
https://www.biorxiv.org/content/early/2018/07/25/374314

We are happy to see so many exciting results coming out of the Open Ephys community. If you are aware of other preprints or publications using Open Ephys, please let us know!

Until next time, 
The Open Ephys Team

Share

November 2017 Newsletter

Added on by Josh Siegle.

Open Ephys at SfN

We're not hosting an info session this year, but there are a number of other opportunities to meet with us next week in DC:

On Monday at 9:35 am in room 145B, Jakob Voigts will be giving a presentation on Open Ephys as part of a Minisymposium on Open-Source Hardware for Neuroscience. Other presenters in this session include Josh Sanders of Sanworks and Gonçalo Lopes of UCL, two scientists/engineers that have worked closely with Open Ephys in the past. We highly recommend attending if you can!

In the Tuesday morning session (VV58), Jon Newman will be presenting a poster on a versatile 64- and 256-channel headstage optimized for closed-loop feedback experiments in freely behaving animals. The headstage plugs directly into a PCIe card, meaning no external acquisition system is required. We plan on manufacturing and selling these headstages via our partnership with OEPS, in order to make them more accessible to the community.

On Monday evening from 7-8:30 pm in room 147A, you can find Josh Siegle at the Neuropixels satellite event. Neuropixels probes, which were funded and tested by HHMI Janelia, Allen Institute, Wellcome Trust, the Gatsby Foundation, and imec, make it possible to record spikes from thousands of cells distributed across dozens of brain regions. They are likely to become a new standard for systems neuroscience research. Neuropixels probes can already be used with the Open Ephys GUI, and we're currently building new plugins to efficiently visualize the data they generate.

After the Neuropixels event, Josh will be DJing the Neurolabware party from 9-10 pm. Stop by the Neurolabware booth at the convention center to pick up your wristband and find out directions to the venue!

Cyclops version 3.6 on sale next week

A new and improved Cyclops LED driver will be available for anyone to order through the Open Ephys store beginning on Monday, November 13th. It's been a long time since we've had this in stock, so we're very excited to start selling it again. These units come fully assembled by OEPS—we're no longer selling kits. The base price is 450 euros, with the option to add an M8 connector (required to interface with LEDs from Thorlabs and Doric Lenses) for an additional 50 euros. This is an incredibly low cost, given how the Cyclops' specifications compare to LED drivers from commercial manufacturers.

GUI version 0.4.3 coming soon

A new version of the GUI is currently in the "testing" phase, meaning updated binaries will be available soon.

Some of the most important upgrades include:

  • The ability to record in 4 different data formats. The "binary" format now uses numpy and JSON files for all data except for continuous signals, which are saved as flat arrays of 16-bit integers.
  • A redesigned backend for defining and accessing information about particular channels. This provides greater transparency and flexibility for developers building new plugins.
  • An overhauled LFP Viewer, which includes a built-in spike raster display and new color schemes.
  • An Event Triggered Averaging module, which can display spike histograms time-locked to digital inputs for both multi-unit activity and single units.

Because this version includes so many new features, it would be extremely helpful if users could proivde some feedback prior to the official release. We'd like people to download the code from the testing branch on GitHub and make sure it works with their preferred experimental signal chain. If you're having trouble getting the code to compile, don't hesitate to email info@open-ephys.org for guidance!

Until next time,

The Open Ephys Team

Share

October 2016 Newsletter

Added on by Josh Siegle.

Announcing Open Ephys Production Site

We’re excited to announce our partnership with Open Ephys Production Site (OEPS), an independent company that will manufacture and distribute the tools featured on the Open Ephys website. In the past, we’ve produced hardware on an ad hoc basis, and it’s been difficult to keep products in stock. Having a dedicated company in charge of hardware distribution should reduce lead times dramatically. For the next two years, OEPS will be handling orders from our online store, and a portion of the profits will be donated to Open Ephys. By purchasing hardware from OEPS, you will help sustain our work in developing and supporting open-source tools for the neuroscience community.

At the beginning of November, OEPS will begin making Acquisition Boards and shipping them worldwide. Soon thereafter, they will begin manufacturing electrode interface boards for the flexDrive, as well as Cyclops LED drivers.

To place a reservation for one or more Acquisition Boards from OEPS, please send an email to info@oeps.tech indicating the quantity you'd like to order. The boards will cost 2250€ ($2500) each, and will include USB 3.0 data transmission by default. Orders can be placed through our existing store, or via purchase order. If you're at all interested, please get in touch. The number of reservations will be used to determine the quantity of boards in the first manufacturing run, to ensure that they don't run out.

The creation of OEPS is being led by Filipe Carvalho, who has spearheaded several previous acquisition board manufacturing runs, and who has extensive experience assembling and using our tools. We're thrilled to have his help in making it easier for neuroscientists to access open-source hardware!

Google Summer of Code

Our participation in the 2016 Google Summer of Code wrapped up in August. We had two students complete projects:

Kirill Abramov (Zaporizhzhya State Engineering Academy, Ukraine) built a Plugin Generator to simplify the process of creating new data processing plugins for the Open Ephys GUI. Instead of copying and pasting example code, there's now an intuitive graphical application for selecting the type of plugin you want to create (Source, Filter, Sink, File Reader, or Record Engine). You can also use the Plugin Generator to add parameters and lay out your plugin's interface. This will lower the barrier to entry for users wishing to add new features to the GUI, allowing them to focus on the functionality, rather than getting their code to run in the first place.

Ananya Bahadur (Amrita University, India) integrated the Cyclops LED driver into the Open Ephys GUI. On the hardware side, he upgraded the microcontroller on the Cyclops from an 8-bit Atmel AVR (Arduino Leonardo) to a more powerful 32-bit Freescale Cortex-M4 (Teensy 3.2). On the software side, he built a CyclopsStimulator plugin for the GUI that provides a simple way to use neural events (such as spikes or LFP features) to trigger LED-based optogenetic stimulation. These additions will improve the stimulation capabilities of the Open Ephys platform. We hope they will facilitate the next generation of experiments involving closed-loop optogenetics.

Ordering Cyclops LED Driver Kits

The most recent updates to Cyclops have been now incorporated into the hardware design, and the final prototypes are currently being tested by Jon Newman. We're planning to manufacture and distribute another round of Cyclops kits for $250 each. Before we place the order for all the components, we'd like to to gauge the level of community interest. If your lab would like to purchase one or more Cyclops kits, please add your name, email, and quantity requested to this Google spreadsheet. This does not constitute the actual order. We will have a separate ordering process once the kits are available.

Until next time,

The Open Ephys Team

Share

June 2016 Newsletter

Added on by Josh Siegle.

More acquisition boards are on the way

The last round of Open Ephys acquisition boards sold out in less than 10 minutes, much faster than we expected. Fortunately, CircuitHub is currently manufacturing additional boards, which will be ready to ship sometime in July. If you were hoping to purchase boards but weren't able to submit an order in time, you can place a reservation for the next round by sending an email to info@open-ephys.org.

The CircuitHub acquisition board packages will include USB 3.0 data transmission (up to 512 channels), and will ship with two assembled I/O boards. You'll have to purchase headstages and cables separately from Intan Technologies.

Next-generation system development

We've been working on a set of standards that will hopefully form the backbone for the next generation of data acquisition systems. We are optimistic that a very simple common standard for hardware and software interconnects can make future data sources highly inter-operable without sacrificing any performance or flexibility. This means that engineers in academia and industry could focus on making better tools rather than re-inventing the data acquisition interface for each new system. The standard is still being developed, so any and all input is welcome. We encourage everyone to read through the latest white paper and get in touch with suggestions and opinions.

Open Ephys at SWC and TENSS

Four members of the Open Ephys team recently traveled to London to take part in a workshop on next-generation electrode technologies organized by Adam Kampff of the Sainsbury Wellcome Centre for Neural Circuits and Behavior. Following the workshop, we headed to Romania to visit the Transylvanian Experimental Neuroscience Summer School (TENSS). TENSS provides a unique opportunity for students from around the world to learn how to build cutting-edge neuroscience rigs in an idyllic, isolated setting. Over the past 3 years, much of the extracellular electrophysiology data in the course has been collected with the Open Ephys acquisition board.

In the days before the course began, we set up a prototype data acquisition system that transfers data via the PCIe bus. PCIe data transmission can greatly reduce the amount of time it takes for neural signals to reach a computer, compared to either USB or Ethernet. By the time the students arrived at the course, we had a demonstration up and running in which spikes occurring in visual cortex were used to trigger optogenetic activation of neurons in motor cortex, with a delay of less than 100 microseconds. The details of the recording system we used (which is a prototype of the next-generation system mentioned above) can be found in this repository.

New software available for download

A new release of the Open Ephys GUI is now available as pre-compiled binaries for Windows, Mac, and Linux. This release (v0.4.1) includes major performance improvements related to visualization and recording, as well as a greatly simplified installation process on OS X. For even more new features currently under development, check out the plugin-GUI development branch (which needs to be compiled from source).

We'd also like to remind everyone to be diligent about using the GUI's Issues page on GitHub. The more information we have about bugs and crashes, the faster they can get fixed!

Until next time,

The Open Ephys Team

Share

April 2016 Newsletter

Added on by Josh Siegle.

Acquisition board pre-orders

CircuitHub is currently manufacturing 20 Open Ephys acquisition boards, which will be ready to ship in mid-May. We will start taking pre-orders through the Open Ephys web store at 12 pm Eastern time on Thursday, April 28th. We expect there to be high demand, so please place your order as soon as possible and limit your order to 2 boards per lab.

The new acquisition boards will come with USB 3.0 connectivity, which allows up to 512 channels of neural data to be acquired simultaneously. They will also ship with two assembled I/O boards, so you won't need to order those separately. The price will be $2500 per package. Headstages and cables can be purchased from Intan Technologies.

For this round, we can only accept payment via credit card. If you need to pay via purchase order, or you don't place your order in time, the Champalimaud Institute will be manufacturing more acquisition boards later this year. We'll send out an update via our newsletter once we know more about when those will be available.

GUI upgrades

We recently migrated our data acquisition software over to a true plugin architecture, which will make it much simpler to add and share new processing modules. Since that update, we've seen a flurry of changes that will considerably improve performance and enhance the user experience: Click-and-drag channel selection (added by Kirill Abramov, @sept-en) String-based channel selection via Matlab-like array syntax, e.g. "1:2:10" (added by Priyanjit Dey, @priyanjitdey94) Disk writing in a separate thread, to reduce CPU usage when recording many channels simultaneously (added by Aarón Cuevas López, @aacuevas) Faster LFP display (added by Jakob Voigts, @jvoigts) Streamlined build process on OS X (added by Christopher Stawartz, @cstawarz) Software icon (added by Josh Siegle, @jsiegle) All of these upgrades are now available in the "testing" branch of the plugin-GUI repository. If you have experience building the GUI from source, we would love help looking for bugs. After at least 2 weeks of testing, we'll move the changes to the "master" branch and make the upgrades available in the pre-compiled binaries.

If you haven't done so already, you should begin merging any changes you've made to the GUI into the plugin-based version. We will no longer be supporting the non-plugin GUI, although we'll leave the repository up if you need to refer to it. If you have any questions about migrating, don't hesitate to email info@open-ephys.org or send a message to our mailing list (open-ephys@google-groups.com).

Google Summer of Code 2016

We'd like to welcome three new members to the Open Ephys team: Kirill Abramov, Jonathan Sieber, and Ananya Bahadur. We selected these three students out of an outstanding pool of GSoC applicants. Between May and August, they will help us improve the Open Ephys GUI. Kirill will create a graphical interface for generating new plugins, which will lower the barriers to adding new functionality to the software. Jonathan will make it possible to write real-time data processing algorithms in Julia, as well as interface the GUI with EEG hardware from OpenBCI. Ananya will create a module for interfacing the GUI with the Cyclops LED driver, to simplify the process of setting up closed-loop experiments. These are features we've been hoping to add for a long time, so we're excited to see what Kirill, Jonathan, and Ananya come up with.

Until next time,

The Open Ephys Team