Skip to content

2026-dhmc-ieeg: NWB/NeuroConv + Neurosift event-related slides #27

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
CodyCBakerPhD merged 4 commits into from
Oct 8, 2026
Merged

2026-dhmc-ieeg: NWB/NeuroConv + Neurosift event-related slides #27

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1c4072e
2026-dhmc-ieeg: NWB/NeuroConv + Neurosift event-related slides
claude Oct 7, 2026
09fa08d
Merge branch 'enh-2026-dhmc-ieeg' into claude/festive-darwin-hvuqwk
CodyCBakerPhD Oct 7, 2026
8875992
Merge branch 'enh-2026-dhmc-ieeg' into claude/festive-darwin-hvuqwk
CodyCBakerPhD Oct 8, 2026
fe6e0f4
Merge enh-2026-dhmc-ieeg (ecosystem-figure restructuring) into PR 27
claude Oct 8, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
145 changes: 122 additions & 23 deletions 2026-dhmc-ieeg.html
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -342,44 +342,110 @@ <h3>Federally regulated but locally &ldquo;managed&rdquo;</h3>
</section>

<section data-transition="slide">
<!-- TODO (Cody): this slide is a skeleton. Please replace/expand
with your own NWB intro framing. -->
<h2>NWB: Neurodata Without Borders</h2>
<ul>
<li>Open standard for neurophysiology data (spikes, LFP, behavior, imaging, ...)</li>
<li>Self-describing, HDF5- or Zarr-backed container format</li>
<li><code>neuroconv</code> &mdash; converts many acquisition-system formats into NWB
(a concrete bridge from clinical/proprietary hardware formats)</li>
</ul>
<aside class="notes">TODO (Cody): swap in your own NWB intro framing.</aside>
<h2>NWB: one file format for all modalities + metadata</h2>
<img data-src="pics/2026-dhmc-ieeg-nwb-one-for-all.svg" style="width:1250px;"/>
<aside class="notes">
Neurodata Without Borders: open standard for neurophysiology data
(spikes, LFP, behavior, imaging, ...) in a single self-describing,
HDF5- or Zarr-backed file &mdash; data and metadata travel together.
</aside>
</section>

<section>
<img data-src="pics/2026-dhmc-ieeg-nwb-convert-scale.svg" style="width:1250px;"/>
<aside class="notes">
Pick the point on the automated &harr; flexible scale that fits:
GUIDE (no code, GUI), NeuroConv (low code, common formats),
PyNWB / MatNWB (custom code against the NWB APIs).
</aside>
</section>

<section>
<img data-src="pics/2026-dhmc-ieeg-neuroconv-formats.svg" style="width:1250px;"/>
<p style="font-size:0.6em;">
<a href="https://neuroconv.readthedocs.io/" target="_blank">neuroconv.readthedocs.io</a>
</p>
<aside class="notes">
NeuroConv format coverage &mdash; a concrete bridge from
clinical/proprietary acquisition hardware (Blackrock, Neuralynx,
EDF, ...) into NWB. Greyed-out entries are in progress.
</aside>
</section>
</section>
<!-- ---- end NWB detour ---- -->


<!-- ---- neurosift detour, triggered by the nwb-apps reveal ---- -->
<!-- light, single-slide detour -- no bookend repeat, straight into
the next increment (archives), same treatment as the NWB section
above. -->
<!-- intro + three example screenshots -- no bookend repeat, straight
into the next increment (archives), same treatment as the NWB
section above. -->
<section>
<section>
<object type="image/svg+xml" data="pics/academic-participants_tuned/nwb-apps.svg" style="width:1300px;height:731px;"></object>
</section>

<section data-transition="slide">
<!-- TODO (Cody): record GIFs of your LFP/spectrogram plugin demo. -->
<h2>neurosift: stream, don't download</h2>
<!-- TODO (Cody): spectrogram plugin example still to come. -->
<h2>Neurosift: stream, don't download</h2>
<ul>
<li>Browser-based NWB viewer &mdash; no download required, works directly against DANDI</li>
<li>Open-source</li>
<li><strong>TODO (Cody)</strong>: record GIFs of your LFP (event-related) and spectrogram plugins &mdash;
live demo won't work over this group's Zoom, needs to be pre-recorded</li>
<li>New <strong>event-related signals</strong> view: align any time series to any column
of a trials/intervals table, pick channels, group trials by a condition</li>
</ul>
<p>
One known iEEG NWB example on DANDI:
<a href="https://openneuro.org/datasets/ds008610/versions/1.0.0/file-display/sub-01:ses-ieeg01:ieeg:sub-01_ses-ieeg01_task-FUS_acq-VPL_run-005_ieeg.nwb" target="_blank">ds008610 (LFP only)</a>
&mdash; TODO: find/convert a second example with spectrograms if one exists.
</p>
<aside class="notes">
Live demo won't work well over this group's Zoom &mdash; the next three
slides are screenshots; each title links to the live view.
</aside>
</section>

<section>
<h3><a href="https://event-related-signals.neurosift.pages.dev/nwb?url=https%3A%2F%2Fapi.dandiarchive.org%2Fapi%2Fassets%2F92e831a1-3427-4e06-9934-30d202bcc02d%2Fdownload%2F&amp;dandisetId=000409&amp;dandisetVersion=0.260309.1324&amp;tab=view%3AEventRelatedSignal%7C%2Fintervals%2Ftrials%5E%2Fprocessing%2Fmotion_energy%2FLeftCameraMotionEnergy#ta_series=%2Fprocessing%2Fmotion_energy%2FLeftCameraMotionEnergy&amp;ta_groupby=gabor_stimulus_contrast&amp;ta_align=auditory_cue_time&amp;ta_win=-4%2C4" target="_blank">Example 1: behavior, grouped by condition</a></h3>
<img data-src="pics/neurosift-event-related-motion-energy-20261007.png" style="width:1300px;"/>
<p style="font-size:0.6em;">
Camera motion energy, aligned to <code>auditory_cue_time</code>,
grouped by <code>gabor_stimulus_contrast</code>
&mdash; <a href="https://dandiarchive.org/dandiset/000409" target="_blank">DANDI:000409</a>
</p>
<aside class="notes">
Solid lines are per-group means, dashed are spread, faint lines are
individual trials. Motion energy dips into the cue, then rebounds
after it &mdash; and the rebound separates by stimulus contrast.
Everything here is streamed from DANDI; nothing downloaded.
</aside>
</section>

<section>
<h3><a href="https://event-related-signals.neurosift.pages.dev/nwb?url=https%3A%2F%2Fapi.dandiarchive.org%2Fapi%2Fassets%2F7a2d22aa-b7c3-43e6-a1e0-129a37bf8ffc%2Fdownload%2F&amp;dandisetId=001539&amp;dandisetVersion=0.250815.1203&amp;tab=view%3ATimeAlignedSeries%7C%2Fintervals%2Ftrials%5E%2Fprocessing%2Fbehavior%2FPosition%2FSpatialSeries#ta_series=%2Fprocessing%2Fecephys%2FLFP%2FElectricalSeries&amp;ta_align=reward_start_time&amp;ta_channels=0%2C2%2C4" target="_blank">Example 2: LFP aligned to reward</a></h3>
<img data-src="pics/neurosift-event-related-lfp-reward-20261007.png" style="width:1300px;"/>
<p style="font-size:0.6em;">
LFP aligned to <code>reward_start_time</code>, channels 0, 2, 4
&mdash; <a href="https://dandiarchive.org/dandiset/001539" target="_blank">DANDI:001539</a>
</p>
<aside class="notes">
Trials without a reward time are skipped and reported in the title
("68 of 95 intervals; 27 without reward_start_time"). Clear evoked
deflection right after reward onset on every selected channel.
</aside>
</section>

<section>
<h3><a href="https://event-related-signals.neurosift.pages.dev/nwb?url=https%3A%2F%2Fapi.dandiarchive.org%2Fapi%2Fassets%2Ffe5b6b44-12a6-4f59-a91f-953850564fb0%2Fdownload%2F&amp;dandisetId=001637&amp;dandisetVersion=draft&amp;tab=view%3ATimeAlignedSeries%7C%2Fintervals%2F40+hz+pulse+train_presentations%5E%2Facquisition%2Fraw_running_wheel_rotation#ta_series=%2Fprocessing%2Fecephys%2FLFP%2FElectricalSeriesProbeD-LFP&amp;ta_channels=5%2C22%2C25%2C32" target="_blank">Example 3: LFP around stimulus presentations</a></h3>
<img data-src="pics/neurosift-event-related-lfp-40hz-20261007.png" style="width:1300px;"/>
<p style="font-size:0.6em;">
40&nbsp;Hz pulse-train presentations, probe&nbsp;D LFP, channels 5, 22, 25, 32
&mdash; <a href="https://dandiarchive.org/dandiset/001637" target="_blank">DANDI:001637</a>
</p>
<aside class="notes">
Any intervals table works, not just "trials" &mdash; here the
stimulus-presentation table. Transient at pulse-train onset, then a
slower negative deflection around 1&nbsp;s.
</aside>
</section>
</section>
<!-- ---- end neurosift detour ---- -->
Expand Down Expand Up @@ -540,7 +606,7 @@ <h2>neurosift: stream, don't download</h2>


<!-- ---- quiet connector nodes: no content of their own yet beyond a -->
<!-- ---- single TODO bullet for dandi-compute -- just flow through -->
<!-- ---- DANDI Compute slides after the dandi-compute reveal -->
<section>
<section>
<!-- TODO: no dedicated nwb2bids slide yet; the tool is already named
Expand All @@ -553,12 +619,45 @@ <h2>neurosift: stream, don't download</h2>
<object type="image/svg+xml" data="pics/academic-participants_tuned/dandi-compute.svg" style="width:1300px;height:731px;"></object>
</section>

<!-- The DANDI Compute figure is a tall poster (1680x2421); the first slide
shows it whole, the following ones crop the same SVG (scaled to
1300px wide, i.e. x0.774) to one panel each. -->
<section data-transition="slide">
<h2>dandi-compute</h2>
<ul>
<li><strong>TODO</strong>: dandi-compute demo &mdash; not yet documented anywhere we could pull from;
need a pointer/screenshot from whoever owns it</li>
</ul>
<style>
.dc-crop { width:1300px; overflow:hidden; margin:0 auto; }
.reveal .dc-crop img { width:1300px; max-width:none; max-height:none; margin:0; border:none; box-shadow:none; }
</style>
<h3>DANDI Compute</h3>
<img class="r-stretch" data-src="pics/dandi-compute-figure.svg"/>
<aside class="notes">
Runs the Allen Institute for Neural Dynamics electrophysiology
pipeline over NWB data on DANDI. Spike-sorting outputs are packaged
as nested BIDS-derivative units; each leaf carries the exact code,
runtime logs, outputs, and provenance metadata needed to re-execute
it &mdash; STAMPED end-to-end (see the STAMPED slide).
Overview only &mdash; the next slides zoom into each panel.
</aside>
</section>

<section>
<h3>DANDI Compute: orchestration</h3>
<div class="dc-crop" style="height:790px;">
<img data-src="pics/dandi-compute-figure.svg"/>
</div>
</section>

<section>
<h3>DANDI Compute: the spike-sorting pipeline</h3>
<div class="dc-crop" style="height:360px;">
<img data-src="pics/dandi-compute-figure.svg" style="margin-top:-822px;"/>
</div>
</section>

<section>
<h3>DANDI Compute: the derivative it produces</h3>
<div class="dc-crop" style="height:580px;">
<img data-src="pics/dandi-compute-figure.svg" style="margin-top:-1216px;"/>
</div>
</section>

<section>
Expand Down
Loading