Polish polar+ring figure: arrows/asterisks swap, un-mirror ring, AR overflow handling#10
Open
mbreiser wants to merge 4 commits into
Open
Polish polar+ring figure: arrows/asterisks swap, un-mirror ring, AR overflow handling#10mbreiser wants to merge 4 commits into
mbreiser wants to merge 4 commits into
Conversation
…verflow handling - Add opts.preview_ring_only flag for fast iteration on just the polar+ring panel - Self-bootstrap src/ onto MATLAB path so load_protocol2_data and friends resolve in a clean session (previously: silent voltage-extraction failure → empty Panel H) - Move direction arrows from outer ring corners onto the polar plot itself, aligned with the 16 grid spokes; filled triangular arrowheads with a small whitespace gap from the 30 mV ring - Move per-direction significance asterisks from the polar plot onto the ring traces, anchored at right-edge / -63 mV inside each subplot — each asterisk now clearly belongs to its trace - Un-mirror the ring layout (drop 180-angle_deg transform) so a direction's polar position and its ring-trace position agree; PD stays at 12 o'clock, ND at 6 o'clock, ODs swap east/west to match the polar - Tighten subplots: subW 0.048 → 0.0433 (~10% narrower) for more inter-trace whitespace; polar_scale 0.65 → 0.70 so the polar fills a bit more of the inner space - Boxplot scatter dots: size 20 → 12 (less visual weight) - AR plot: clamp any value above the y-cap to y_top and draw a small upward triangle above to indicate off-axis (preserves the data point at AR=4.573 in the 28 dps T4 ctrl group without breaking the new y-axis cap at 4); shift significance brackets up to clear the triangle Verification: both 56 dps and 28 dps render end-to-end; voltage extraction succeeds (48/48 cells); asterisk placement on the ring now matches the polar plot's ctrl-vs-tutl- gaps on the same side. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
… speeds The polar direction arrows were anchored to rL(2)=30 mV (data coords), which meant their visual position relative to the polar box edge varied with opts.rpad_factor. On 56 dps (rpad_factor=1.10) the arrows landed cleanly outside the box; on 28 dps (rpad_factor=1.15) the same data positions ended up inside the box, overlapping the polar grid. Switch r_tail / r_head to be multiples of rpad (the actual axFill box edge). The 56 dps appearance is essentially unchanged (33.6→33.66 mV tail). The 28 dps arrows move outward by ~5% and now sit clearly outside the polar grid like on the main figure. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
…e caps)
MATLAB Line objects don't expose a LineCap property, so the LineWidth=2.5
horizontal FWHM bars were rendered in PDF with the default projecting
('square') cap — extending ~1.25 pt beyond each geometric endpoint and
making the visible bar appear ~2.5 pt longer than the actual FWHM extent.
Replace the two plot() calls in add_fwhm_bars with patch() rectangles
whose width is exactly rx-lx in data coords. Bar thickness is computed
from the axes' Position in points so it matches the prior visual ~2.5 pt
height. The endpoint tick marks remain as thin plot() lines (their cap
projection is negligible at LineWidth=1.0).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
… figure)
When opts.rotate_pd_left = true, the polar+ring panel is rotated CCW by
90° so PD points LEFT (and ND points RIGHT), matching the orientation of
the M6-aligned PD-axis flash figure (generate_manuscript_fig_ef.m with
axis_mode='pd' produces PD on the left side).
Implementation: a single rot_offset (0 or pi/2) is threaded through
draw_ring_panel and plot_polar_with_patch. It is added to:
- the ring subplot angle (positions all 16 time-series around the ring)
- the theta argument to draw_polar_patch (rotates ctrl/tutl- curves)
- the theta inputs to pol2cart for the polar arrows
- the 30 mV label position
- the PD/OD/ND reference lines (via a rot2d helper) and labels
(with HorizontalAlignment/VerticalAlignment swapped so labels stay
on the outer side of the tip)
Ring asterisks, scale bar, and per-axes drawing are unaffected — they
follow whichever subplot they belong to.
Outputs are saved with a _rot90 suffix so they coexist with the standard
(PD-up) outputs. Other panels (boxplots, AR, V_m) are unchanged.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Cosmetic polish + small infrastructure fixes to
generate_manuscript_fig_ds.m. Affects both 56 dps (main figure) and 28 dps (supp), specifically the polar+ring-of-traces panels and the right-side boxplot/AR panels.Changes
Layout / cosmetics
180 − angle_degtransform so a direction's polar position and its ring-trace position agree. PD stays at 12, ND at 6; the ODs swap east/west to match the polar. This addresses a long-standing visual inconsistency that became obvious once asterisks lived on the ring.subW0.048 → 0.0433) for more inter-trace whitespacepolar_scale0.65 → 0.70 so the polar fills more of the inner spaceAR plot overflow handling
y_limits = [0 4.4]cap are drawn aty_top = 4with a small upward filled triangle above to signal "off-axis high"Infrastructure
opts.preview_ring_onlyflag added — renders only Panels A & B (polar + ring), writes to a_preview_ringfilename. Used as a fast iteration loop on the polar+ring panel.src/onto the MATLAB path relative to its own file location, soload_protocol2_data/parse_bar_data/parse_bar_data_pre_bfresolve in a fresh session. Previously the function would silently fail voltage extraction and emit an empty Panel D / H whensrc/wasn't already on the path.Test plan
generate_manuscript_fig_ds(56)renders all panels including populated V_m boxplotsgenerate_manuscript_fig_ds(28)renders all panels with AR=4.573 outlier shown as capped marker + up-arrowgenerate_manuscript_fig_ds(56, struct('preview_ring_only', true))renders just the polar+ring as*_preview_ring_*.pdf/.pngscripts/on path) — function bootstrapssrc/itself🤖 Generated with Claude Code