Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data

arXiv:2611.15606 · q-bio.NC · Submitted 2026-11-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Genomics Radio. Generated commentary on the latest computational biology and genomics papers.

Ines: We're looking at the paper: "Optogenetic Modulation Fraction".

Marcus: Here's the gist of the paper. The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>.

Ines: Pruning redundant reasoning tokens The paper reanalyzes eight public Allen Institute Visual Coding Neuropixels sessions (4,112 units) to rigorously test the effect of expressing channelrhodopsin-2 (ChR2) in interneuron populations on spike timing alteration under light Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>.

Marcus: The analysis involves modeling latency from light-pulse onset to first spike as a right-censored time-to-event outcome using a Cox proportional-hazards model with a single BenjaminiHochberg FDR correction pooled across all units Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>.

Ines: The study reports two complementary quantities: an unrestricted modulation fraction (any unit with an altered spike-timing hazard) and a restricted direct-responder fraction (modulated units whose light-evoked latency is under 10 ms) Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>.

Title and authors: Ines: We’ve covered the main title and authors, so let’s talk about what that title actually tells us about the research. The paper is "Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data".

Marcus: It points out two things immediately: first, there’s this robust effect of opsin presence on modulation fraction, and second, they admit that the cell type itself is still an unresolved question. That means they found a strong overall signal but couldn't fully nail down which specific neuron was responsible for all the changes >

Yuki: That’s important because it tells us that while we can manipulate a population of interneurons with light and see a general effect, we still don't know for sure if it’s because of one specific subtype responding better than others >

Ines: It frames the whole thing as testing an assumption that is usually never directly tested: whether light reliably produces the intended effect in genetically targeted populations. Altahini, Arnoux, and Stroh’s review points this out by noting most circuit-dissection studies don't measure what fraction of the targeted population actually responds one > <ref:2611.15606#pg2,what fraction of the targeted population actually>

Marcus: So the paper is trying to go beyond just saying light worked; they are quantifying exactly what fraction of the whole local population was affected by stimulating that targeted group, which is a different thing than just asking if the specific target responded >

Yuki: That distinction is key because it separates whether we’re looking at what fraction of the targeted population responded versus what fraction of the whole local population was affected by stimulating that targeted population >

Ines: And that distinction they address directly in Section three point four by restricting their analysis to a latency-based criterion for putative direct responses, rather than treating the unrestricted, population-level fraction as a stand-in for targeted-population response > <ref:2611.15606#pg3,to a latency-based criterion for putative direct responses, rather than treating>

Marcus: It’s like they’re trying to be more precise about what that modulation fraction actually means in terms of biology versus just physics of light hitting cells >

Yuki: And that precision helps us connect this finding to the wider history of the species, because if we can't definitively say which cell type is driving the signal, it complicates how we interpret evolutionary pressures >

The paper's summary: Ines: So putting it together, what’s the actual summary of what they did and what they found in "Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data"?

Marcus: Basically, they reanalyzed eight public sessions from the Allen Institute Visual Coding Neuropixels dataset. They used a right-censored time-to-event outcome to model the latency from light pulse onset to the first spike >

Yuki: And they used a Cox proportional hazards model with log-rank fallback, adding a single Benjamini Hochberg FDR correction pooled across all four thousand one hundred twelve units in their analysis > <ref:2611.15606#pg1,a Cox proportional hazards model>

Ines: The main summary is that expressing ChR2 in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls under every method and outcome definition they tried >

Marcus: They report two complementary quantities: the unrestricted modulation fraction, which is any unit with an altered spike-timing hazard, and a restricted direct-responder fraction, which only counts modulated units whose latency is under ten milliseconds >

Yuki: The unrestricted fraction showed a large elevation over control under both metrics—zero point two three to zero point three seven versus the control's zero point zero one to zero point zero five after FDR correction—and their formal genotype test confirmed that Sst or Pvalb sessions showed a significant elevation, p value of zero point zero three one two >

Ines: That’s a big number for us; it means the manipulation is working broadly across the population, not just hitting one tiny subset of cells >

Marcus: But as we looked at the restricted fraction, they saw that while the ordering between control and Vip and Sst or Pvalb was preserved, that gap between Vip and control narrowed considerably in relative terms >

Yuki: And they also found a descriptive pattern suggesting a reversal—that among modulated units, Vip sessions had a higher proportion of putative direct responses at thirty-four percent than Sst or Pvalb sessions which were around twenty to twenty-eight percent >

Ines: But as we discussed, that descriptive pattern didn't survive the same formal cluster-robust bootstrap test applied to it, because the true no-opsin control group actually had the highest direct-response proportion of all three groups at forty-two point six percent >

The paper's improvements: Marcus: Now, let’s talk about what they suggest is improved in this paper and what that means for how we approach these kinds of studies. The authors are pointing out that the main methodological lesson is that pooled estimates, standard cluster-robust inference, and descriptive secondary patterns in small-sample circuit-dissection studies all deserve the same formal scrutiny as a paper’s headline claim >

Ines: They suggest a four-correction paper approach because they took that into account: they considered a composition-dependent artifact in a smaller session pool, and the standard versus exact bootstrap discrepancy described above >

Yuki: The biggest improvement suggested is using the exact wild cluster bootstrap when assessing genotype effects across limited session clusters to avoid over-rejecting results from just standard cluster robust logistic regression >

Marcus: So they want us to be more careful about relying on that one statistical test when we look at genetic differences because that test might be too strong for the job >

Ines: It’s a shift toward demanding that any descriptive secondary pattern you see in a paper, even if it looks like a finding, must pass the same rigorous small-cluster methods applied to the main claim >

Yuki: That level of scrutiny is what we need when we’re trying to connect findings across different species or cohorts because it prevents us from just accepting descriptive observations without formal testing >

Conclusion: Ines: So, wrapping up the paper "Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data". The central conclusion is that ChR2 expression itself substantially elevates modulation fraction over true no-opsin control under every method they tried >

Marcus: They’re arguing that a descriptive, genotype-stratified proportion can look like a finding and not be one until the same formal test applied to a paper’s headline claim is applied to it as well >

Yuki: For me, the implication is that we need to treat those secondary findings with the same statistical weight as the main result when we are trying to build up our understanding of how different cell types respond >

Ines: It means we need to be disciplined about what’s considered a finding versus what's just an interesting observation, especially in circuit dissection studies >

Marcus: And for this paper, it really settles the question that ChR2 expression is a major driver of modulation fraction over control under various metrics >

Yuki: It sets the stage for future work where we can better distinguish between direct and network-mediated effects by using latency thresholds to resolve that cell-type ambiguity >

Ines: Exactly, and it’s a good reminder that we need to be disciplined about what’s considered a finding versus what's just an interesting observation, especially in circuit dissection studies >

Siddhardha Nanda

Columbia University

q-bio.NC

Submitted: 2026-11-11

Updated: 2026-11-11

Comments: 15 pages, 2 figures, 5 tables

Code: https://github.com/py-econometrics/wildboottest

License: http://creativecommons.org/licenses/by/4.0/

Importance score: 83/100

The gist: The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike…

Key concepts

Optogenetic Modulation Fraction
This metric measures the percentage of neurons whose spike timing (latency from light pulse to first spike) is altered when they express ChR2. The study uses two versions: an unrestricted fraction (any unit with altered timing) and a restricted fraction (units with latency under 10 ms). This helps quantify the overall effect of ChR2 expression.
Censoring-Aware Survival Analysis
The researchers used survival analysis to model the time it takes for a neuron to spike after light stimulation. Because some neurons never spiked within a set time window, this method correctly handles 'censored' data—where the outcome is unknown but known not to have occurred yet. This provides a more accurate measure of latency than simple averages.
Genotype Dependence Testing
To determine if the effect of ChR2 expression differs between specific cell types (like Sst/Pvalb vs. Vip), the researchers used formal statistical tests, including cluster-robust logistic regression and an exact wild cluster bootstrap. This rigorous testing helps ensure that observed differences in modulation fraction are statistically significant and not due to random chance within specific session groups.

Terminology

Summary

The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0

Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

Pruning redundant reasoning tokens

The paper reanalyzes eight public Allen Institute Visual Coding Neuropixels sessions (4,112 units) to rigorously test the effect of expressing channelrhodopsin-2 (ChR2) in interneuron populations on spike timing alteration under light Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. The analysis involves modeling latency from light-pulse onset to first spike as a right-censored time-to-event outcome using a Cox proportional-hazards model with a single BenjaminiHochberg FDR correction pooled across all units Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0

Where the paper says it

We reanalyze eight public Allen Institute Visual Coding Neuropixels sessions (4,112 units), modeling the latency from light-pulse onset to first spike as a right-censored time-to-event outcome, testing it with a Cox proportional-hazards model (log-rank fallback) and a single BenjaminiHochberg FDR correction pooled across all units, and assessing genotype dependence with a formal, small-cluster-robust statistical test (standard cluster-robust logistic regression plus an exact wild cluster bootstrap, since only eight session-level clusters are available). Because a Neuropixels probe records the whole local population, not only the genetically targeted cells, we report two complementary quantities: an unrestricted modulation fraction (any unit with an altered spike-timing hazard, direct or network-mediated) and a restricted direct-responder fraction (modulated units whose light-evoked latency is under 10 ms, an explicitly stated threshold, since no universal numeric standard exists in the optotagging literature).

The study reports two complementary quantities: an unrestricted modulation fraction (any unit with an altered spike-timing hazard) and a restricted direct-responder fraction (modulated units whose light-evoked latency is under 10 ms) Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0
Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

How the training signal is built

The survival framing measures latency from pulse onset to the first spike within a fixed 50 ms post-pulse window, treating no spike as right-censored at 50 ms Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. Significance testing employs a Cox proportional-hazards model with a log-rank fallback, followed by a single BenjaminiHochberg FDR correction pooled across all 4,112 units Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0

Where the paper says it

We reanalyze eight public Allen Institute Visual Coding Neuropixels sessions (4,112 units), modeling the latency from light-pulse onset to first spike as a right-censored time-to-event outcome, testing it with a Cox proportional-hazards model (log-rank fallback) and a single BenjaminiHochberg FDR correction pooled across all units, and assessing genotype dependence with a formal, small-cluster-robust statistical test (standard cluster-robust logistic regression plus an exact wild cluster bootstrap, since only eight session-level clusters are available). Because a Neuropixels probe records the whole local population, not only the genetically targeted cells, we report two complementary quantities: an unrestricted modulation fraction (any unit with an altered spike-timing hazard, direct or network-mediated) and a restricted direct-responder fraction (modulated units whose light-evoked latency is under 10 ms, an explicitly stated threshold, since no universal numeric standard exists in the optotagging literature).

To test genotype dependence formally, the authors used a cluster-robust logistic regression on session clusters, supplemented by an exact wild cluster bootstrap on a linear probability model Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0
Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

Where the gains come from

The unrestricted modulation fraction for ChR2-expressing Sst/Pvalb sessions was found to be large, showing an elevation over control under both metrics (unrestricted 0.23-0.37 versus 0. p. 0

Where the paper says it

ChR2-expressing Sst/Pvalb sessions show a large, bootstrap-confirmed elevation over control under both metrics (unrestricted 0. 37 versus 0.

01-0.05 FDR-corrected, bootstrap p = 0. p. 0
Where the paper says it

37 versus 0. 05 FDR-corrected, bootstrap p = 0.

0312; restricted 0. p. 0
Where the paper says it

05 FDR-corrected, bootstrap p = 0. 0312; restricted 0.

077 versus 0. p. 0
Where the paper says it

0312; restricted 0. 077 versus 0.

013 pooled, bootstrap p < 0. p. 0
Where the paper says it

077 versus 0. 013 pooled, bootstrap p < 0.

06) Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0
Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

The formal genotype test revealed that while Sst/Pvalb showed a significant elevation over control (p = 0. p. 7
Where the paper says it

2 and leaves another unchanged. Table 4 shows that Sst/Pvalb’s elevation over control remains significant under both the logistic GLM (p = 0.

0312), the Vip versus control contrast did not survive the exact wild cluster bootstrap (p = 0. p. 7
Where the paper says it

The direct Sst/Pvalb-versus-Vip contrast, however, is not resolved: the logistic GLM remains significant (p = 1. 3×10−4 ) while the exact wild cluster bootstrap gives p = 0.

125) Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0
Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

The analysis of the restricted direct-responder fraction showed that the ordering (control < Vip < Sst/Pvalb) was preserved, but the gap between Vip and control narrowed considerably in relative terms Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. Crucially, a descriptive pattern suggesting a reversal—that among modulated units Vip sessions had a higher proportion of putative direct responses (34%) than Sst/Pvalb sessions (20-28%)—did not survive the same cluster-robust bootstrap test applied to it Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0

Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

This descriptive pattern was explained by the fact that the true no-opsin control group has the highest direct-response proportion of all three groups (42. p. 0
Where the paper says it

Second, a descriptive pattern in the restricted data, that among modulated units Vip sessions had a higher proportion of putative direct responses (34%) than Sst/Pvalb sessions (20-28%), the opposite of the raw ordering, looked like a genuine cell-type finding until we applied the same cluster-robust bootstrap test used everywhere else in this paper to the proportion itself: none of the three genotype contrasts on this metric approach significance (p = 0. 84), and pooled across sessions the true no-opsin control group has the highest direct-response proportion of all three groups (42.

6%), exceeding Vip Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0
Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

The authors conclude that the central, robust finding is that ChR2 expression itself substantially elevates modulation fraction over true no-opsin control under every method and outcome definition tried Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. The paper argues that a descriptive, genotype-stratified proportion can look like a finding and not be one until the same formal test applied to a paper’s headline claim is applied to it as well Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data>. p. 0

Where the paper says it

Optogenetic Modulation Fraction: A Robust Opsin-Presence Effect and an Unresolved Cell-Type Question, by Censoring-Aware Survival Analysis of Public Neuropixels Data Siddhardha Nanda Columbia University sn3199@columbia. edu November 2026 Abstract The clearest result in this paper is simple: expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially and robustly raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method and outcome definition we tried.

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

  1. Improved AI system can perform robust quantification of optogenetic efficacy by using a censoring-aware survival framing and applying a single pooled multiple-comparisons correction across all units, as described in Section 3.2.

  2. The improved system can distinguish between direct and network-mediated responses by restricting analysis to putative direct responders defined by an explicit threshold, such as a median light-evoked latency under 10 ms, which helps resolve the population-versus-targeted-cell ambiguity.

  3. The AI system can employ a formal small-cluster robust statistical test, specifically the exact wild cluster bootstrap, when assessing genotype effects across limited session clusters to avoid over-rejecting results from standard cluster robust logistic regression.

  4. The system can be trained to identify and retract secondary descriptive patterns in circuit dissection studies that appear as genuine findings but fail formal testing with the same rigorous small-cluster methods applied to headline claims, such as the reversal in direct-response proportion among modulated units.

Abstract

Expressing channelrhodopsin-2 (ChR2) in an interneuron population substantially raises the fraction of locally recorded units whose spike timing is altered by light, relative to no-opsin controls, under every method tried. Establishing that result rigorously, and retracting two secondary claims that did not survive the same rigor, is this paper's actual content. We reanalyze eight public Allen Institute Neuropixels sessions (4,112 units), modeling light-to-first-spike latency as a right-censored time-to-event outcome, with a Cox model, pooled FDR correction, and a small-cluster-robust test (logistic regression plus an exact wild cluster bootstrap, since only eight session clusters exist). We report an unrestricted modulation fraction and a restricted direct-responder fraction (evoked latency under 10 ms, an explicit, literature-grounded threshold). ChR2-expressing Sst/Pvalb sessions show a large, bootstrap-confirmed elevation over control under both metrics. Two further claims, each initially suggested by a standard test, did not survive the exact bootstrap or a direct formal check, and we report both retractions as results in their own right. Vip-ChR2's apparent elevation over control is not confirmed by the bootstrap and becomes non-significant once restricted to direct responders. A second pattern, that among modulated units Vip sessions had a higher proportion of direct responses than Sst/Pvalb, opposite the raw ordering, evaporated under the same bootstrap test: no genotype contrast approaches significance, and control has the highest proportion of all three, which cannot be biological. Per-session rates vary enormously within one genotype. With two earlier corrections (a session-composition artifact and a standard-versus-exact-bootstrap discrepancy), this is a four-correction paper: pooled estimates and secondary patterns deserve the same scrutiny as a headline claim.

Related papers