Corpus 1,272 assessed · 1,173 scored · 643 reproduced ≥75 · 168 flagged ·∅ 74.1/100
← New search

Extensive tissue-specific expression variation and novel regulators underlying circadian behavior.

Sci Adv · 2021
L1 63/100 3/4
Why this verdict

The main results reproduced, with only marginal, non-material deviations.

Reproduced on the brainbox compute brainarbeit.com
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +3
✓ What held up
  • Same input data as the authors
  • Reported values were directly comparable
  • The central claim held under reproduction
What did not (or only partly)
  • 🔴A deviation arose in the data or preprocessing
  • 🔴A deviation was attributed to the published material
  • 🟡Reported values were not (fully) derivable from the shared data
  • 🟡The deviation was non-trivial in magnitude
  • 🟡Overall, the reproduction showed a material discrepancy
How its reproducibility compares
63/100
Reproducibility score
0.6 SD below mean
vs. all fields · 1173 studies
🎯 Scores higher than 24% of all assessed papers rank 875 of 1173 scored

A 0–100 reproducibility-quality score from the per-question grades, shown as a z-score: standard deviations above (+) or below (−) the mean of comparable assessments.

Reproduction agent’s raw note

Described well enough to reproduce DOWNSTREAM results faithfully. The paper's processed matrices are deposited on GEO (GSE126018), so we reproduced pipeline-derived results directly on them rather than re-running BRB-seqTools/STAR/HTSeq on 778 SRA runs. EXACT 1:1: expressed-genes-per-sample (6915/13131/9933). EXACT (qualitative): all 7 canonical clock genes cycle in all 4 tissues (JTK_CYCLE), and DGRP-796 loses ~half its rhythmic transcriptome in both brain and gut (Fig 5F). WITHIN-TOL: Fig5F gut counts (358 vs 362; 680 vs 717). PARTIAL/MISMATCH: absolute per-tissue cycling-gene tallies (135/359/479/440) come out ~1.5-2.4x higher with MetaCycle's JTK_CYCLE v3.1 at ADJ.P<0.05 single-24h period - same method, same rank-order (gut>MT>fatbody>brain), same magnitude, but exact JTK settings/extra filtering aren't pinned by the Methods, and the reported per-tissue sum (1413) contradicts the stated total (1757), so those reported anchors need human checking. NOT attempted: 45-DGRP-lines aberrant-circadian (32.1%), CircaCompare phase shift, GENIE3 network. Dataset GSE126018 profiled: open, complete, N=778 matches exactly, grade A. Overall: honest partial reproduction - headline statistic and central biology reproduce; exact cycling counts limited by under-specified parameters.

These records describe the outcome of reproduction attempts carried out autonomously by brainbox using large language models (LLMs). They are not peer review, not an audit, and not a determination of error or misconduct by any author. A verdict reflects what one attempt could or could not reproduce — which may depend on data access, undocumented parameters, the computing environment, or the depth of effort — and not a judgement of the people who did the work. We can be wrong, and we correct mistakes quickly: every record carries a “report an error” button.

Assessment versions

Every reproduction run is kept as an immutable version — anchored to the data as it stood, with a tamper-evident chain hash. A rerun (e.g. after an author updates a deposit) adds a new version; the previous one stays on record.

  1. v1 current initial assessment Score 63
    assessed: 2026-06-18 ⛓ f2134371cff7
✎ I am an author of this paper

Updated or fixed a deposit, or is there an erratum? Ask us to re-run the metrics. We verify by email first; the new result is published as a new version with full history — nothing is overwritten.

Reason for the rerun

We email you a confirmation link first. The rerun is an objective re-measurement — it cannot change the verdict in your favour, only ask us to look again.

Provenance — full disclosure

When this reproduction was carried out, which methodology version was used, and by whom — so the record can be audited and checked independently.

Reproduced
2026-06-18
Rubric version
v1.0
Assessed by
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-18
no human curator yet
Last updated
2026-08-05

Provisional, curator- or AI-assessed, and independently checkable. A reproduction outcome states what one attempt could reproduce — not a judgement of the authors.

Deep full-text extraction

Model: opus
Founding hypothesis

How does natural genetic variation affect circadian rhythms at the molecular level, and how does the circadian clock control tissue-specific gene expression rhythms across genotypes? The paper tests this by generating tissue-specific transcriptomes of D. melanogaster to characterize population-level molecular circadian clock variation.

Core claims
  • More than 1700 mostly tissue-specific cycling genes are detected across four Drosophila tissues, substantially expanding the catalog of clock-controlled genes finding
  • Seven previously uncharacterized genes (CG2277, CG5793, CG31324, CG14688, Gclm, Amph, Usp1) cycle in all tissues and represent novel putative circadian regulators finding
  • Over 30% (45/141) of sampled DGRP lines exhibit aberrant, tissue-specific circadian gene expression, revealing intertissue circadian expression desynchrony driven by genetic variation finding
  • A novel cry mutation in DGRP-796 disrupts light-driven FAD cofactor photoreduction, providing in vivo support for the conserved photoentrainment mechanism mechanism
  • A regulatory TF hierarchy with tissue-specific cycling TFs acts downstream of master clock regulators to generate local gene expression rhythms mechanism
  • Glutathione metabolism is under tissue-specific, desynchronized circadian control, the first reported naturally occurring tissue-specific circadian molecular desynchrony of a biological process in D. melanogaster finding
  • A comprehensive tissue- and genotype-specific circadian gene expression atlas (>700 transcriptomes) was generated as a community resource resource
  • RNAi knockdown of Amph, CG2277, CG5793, and Usp1 in clock neurons produces consistent locomotor rhythm defects, validating their role in the pacemaker finding
Experimental setups
Assay System Perturbation Readout Platform
bulk RNA-seq (BRB-seq, 3'end counting) D. melanogaster w1118 reference strain; brain, gut, Malpighian tubules, fat body none (12h/12h LD followed by 24h DD at 25°C) genome-wide gene expression periodicity/rhythmicity (cycling genes) BRB-seq
bulk RNA-seq (high temporal resolution RNA-seq) 141 DGRP lines; three tissues none (genetic variation across lines) circadian gene expression / aberrant cycling
locomotor activity rhythm assay (behavioral perturbation) Drosophila clock cells (tim-GAL4) and Pdf-expressing LNv central clock neurons (Pdf-GAL4) UAS-RNAi knockdown of candidate genes % rhythmic flies, period, rhythmicity strength/index, activity levels UAS-RNAi constructs (Bloomington and VDRC stocks)
gene regulatory network / ChIP-seq integration Drosophila cycling genes across four tissues; Clk/Cyc binding data none network connectivity, direct Clk/Cyc targets, TF modules Cytoscape; publicly available Clk ChIP-seq data
intertissue phase shift analysis (CircaCompare) genes cycling in two or more Drosophila tissues none phase shift in expression between tissue pairs CircaCompare
protein structural modeling and simulation Cry protein (DGRP-796 mutation) novel cry mutation effect on light-driven FAD cofactor photoreduction
brain immunohistochemistry DGRP-796 Drosophila brain novel cry mutation clock function in brain
comparative ortholog circadian expression analysis mouse (Mus musculus) and baboon (Papio anubis) tissues none cycling of mammalian orthologs across tissues public databases / prior publications
Key results
  • 1757 cycling genes detected across four tissues, most cycling tissue-specifically 1757 genes
  • Only 14 genes cycled in all four tissues, including 7 canonical clock genes and 7 uncharacterized genes 14 genes
  • 45 of 141 (>30%) DGRP lines showed aberrant circadian gene expression >30% (45/141)
  • 344 genes (19.6%) cycled in two or more tissues; 101 genes (29%) showed phase shift >2 hours between tissues 344 genes (19.6%); 101 genes (29%)
  • Glutathione-associated genes were delayed in fat body relative to gut/Malpighian tubules phase shift 4.5 to 7.8 hours
  • Knockdown of Amph, CG2277, and Usp1 increased oscillation period and decreased rhythmicity index period 23.8–25.1 hours
  • 52% of TSC genes directly connected to core regulators (Clk, cwo, vri, Pdp1); 20% directly bound by Clk/Cyc 52%; 20%
  • Most tissue-specific cycling TFs (37 of 38) showed tissue-specific cycling but not tissue-specific expression 37 of 38
Key statistics
  • count 233 transcriptomes (w1118 reference) (four tissues profiled every 2 hours over 48 hours in triplicate)
  • count 451 transcriptomes (DGRP screen) (141 DGRP lines across three tissues)
  • count 1757 cycling genes (detected across four tissues by JTK_CYCLE)
  • fold_change 45 (>30%) of 141 DGRP lines (lines with aberrant circadian gene expression)
  • pvalue P = 5.1 × 10^−5 (GO enrichment glutathione metabolic process (Fisher's exact test))
  • pvalue P = 2.2 × 10^−3 (GO enrichment glutathione transferase activity (Fisher's exact test))
  • count 101 genes, 29%, FDR < 0.05 (phase shift >2 hours between at least one pair of tissues)
  • other phase shift 4.5 to 7.8 hours (glutathione gene delay in fat body vs gut/Malpighian tubules)

Statistical methods review

Model: sonnet

A neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.

The study generated >700 tissue-specific transcriptomes from Drosophila melanogaster (w1118 reference strain sampled every 2 hours over 48 hours in triplicate, plus 141 DGRP lines) and identified cycling genes using JTK CYCLE. Intertissue phase shifts were assessed with CircaCompare; GO enrichment was tested with Fisher's exact test; and knockdown locomotor behavior was evaluated using a test of equal proportions (% rhythmic flies) and ANOVA (period length and rhythmicity index). Multiple-testing correction was applied throughout and reported as FDR-adjusted P values with significance indicated by star notation.

Replicationbiological Sample sizeTriplicates per time point for w1118 reference strain; 141 DGRP lines for population screen; total 233 (w1118) + 451 (DGRP) transcriptomes stated; no explicit power analysis described Groupsw1118 reference vs. 141 DGRP lines across brain, gut, and Malpighian tubules; RNAi knockdown lines vs. tim-GAL4 or Pdf-GAL4 controls for locomotor behavior; four tissues compared for cycling and phase Pairingunpaired Randomization/blindingnot stated Dispersionnone Exact p-valuesno Effect sizesno Confidence intervalsno Multiplicity correctionFDR (specific algorithm, e.g. Benjamini-Hochberg, not named in the available text)
Statistical tests used
Test Applied to n Assumptions
JTK CYCLE (nonparametric rhythmicity detection) Identification of cycling genes across four tissues (brain, gut, Malpighian tubules, fat body) in w1118 reference time series 233 transcriptomes from w1118 reference strain (4 tissues × 2-hour sampling × 48 hours × triplicates) not stated
CircaCompare (phase shift comparison) Intertissue phase shift analysis of genes cycling in two or more tissues 344 genes cycling in ≥2 tissues not stated
Fisher's exact test GO enrichment analysis of phase-shifted genes (glutathione metabolic process P=5.1×10⁻⁵; glutathione transferase activity P=2.2×10⁻³) not stated
Test of equal proportions Comparison of % rhythmic flies between RNAi knockdown lines and controls in locomotor behavior assays (Fig. 2) Per-gene N reported in Fig. 2 column not stated
ANOVA Comparison of locomotor period length and rhythmicity index across RNAi knockdown lines and controls (Fig. 2) Per-gene N reported in Fig. 2 column not stated
FDR-adjusted significance testing (specific algorithm not named) GO enrichment of TSC genes (adjusted P values reported); intertissue phase shift analysis (FDR<0.05 threshold for 101 genes); locomotor behavior comparisons (FDR-adjusted P in Fig. 2) not stated
Approaches that could also have been used
  • JTK CYCLE alone was used to classify genes as rhythmically expressed across tissues
    Could also: RAIN (Rhythmicity Analysis Incorporating Non-parametric methods) or harmonic/cosinor regression (e.g., via R packages HarmonicRegress or rain) could also be applied to the same time-series RNA-seq data — Different rhythmicity algorithms differ in sensitivity and specificity depending on time-series length and sampling regularity; reporting concordant hits across two methods is an approach used in comparable circadian transcriptomics studies to characterize the robustness of the cycling gene list
  • ANOVA was used to compare period length and rhythmicity index across multiple RNAi lines and controls, with FDR adjustment applied
    Could also: A one-way ANOVA followed by a named post-hoc procedure — Dunnett's test (each knockdown vs. a single control) or Tukey HSD (all pairwise) — could also be applied — Named post-hoc procedures yield per-comparison adjusted P values and make explicit which specific RNAi lines differ from control, which aids interpretation when several lines are tested per gene with two different drivers
  • FDR correction was applied across multiple test families, but the specific algorithm was not named
    Could also: Explicitly naming the FDR procedure (e.g., Benjamini-Hochberg 1995 or Storey q-value) and reporting q-values alongside raw P values would also be standard practice — BH-FDR and the Storey q-value differ in conservatism and power; naming the method lets readers assess stringency and reproduce the threshold-setting step exactly
  • Fisher's exact test was used for GO enrichment of phase-shifted and TSC gene sets
    Could also: Permutation-based enrichment testing or a topology-aware GO analysis (e.g., topGO 'weight' or 'elim' algorithms in R) could also be applied — Fisher's exact test treats each GO term independently; topology-aware methods account for the hierarchical dependency structure of the GO DAG and can reduce redundancy among enriched ancestor and descendant terms
  • The percentage of rhythmic flies was compared between knockdown and control groups using a test of equal proportions
    Could also: A chi-square test or logistic regression with group as predictor and rhythmicity as the binary outcome could also be applied — Logistic regression additionally accommodates covariates (e.g., RNAi construct identity, driver line) and provides odds ratios as a quantified effect size alongside significance, which complements the star-notation summary shown in Fig. 2
  • Sample size for the w1118 reference time series was set to triplicates per time point without a stated power analysis
    Could also: An a priori power calculation for time-series rhythmicity detection — for example via simulation or tools such as the R package SSPA — could also be reported alongside the design — Stating the expected effect size (e.g., amplitude of oscillation), variance, and resulting power at the chosen FDR threshold helps readers evaluate whether the design was sized to detect low-amplitude cycling genes, which are often the biologically relevant but harder-to-detect class
Software: JTK CYCLE · CircaCompare · Cytoscape · BRB-seq (bulk RNA barcoding; sequencing library method)

What was reproduced

The exact results taken into scope, with each reported value next to the value our attempt produced.

Scope — pmid-33514540 (Litovchenko et al., Sci Adv 2021; GSE126018)

In scope (pipeline-derived, attempted)

  • Expressed-genes summary (count>0 per sample) from the deposited raw count matrix. Pipeline: simple matrix summary. → reproduced EXACT.
  • Cycling-gene detection per tissue via JTK_CYCLE v3.1 (Methods explicitly names it) on the deposited voom+ComBat normalized per-tissue matrices. Tool used: MetaCycle 1.2.0 meta2d(cycMethod="JTK"), which bundles JTK_CYCLE v3.1.
    • w1118 reference, 4 tissues (brain, gut, fat body, Malpighian tubules).
    • Fig 5F matched w1118 vs DGRP-796 (brain, gut) from the combined matrices.
    • Core (intersection across 4 tissues) + main clock-gene identity.

In scope but NOT attempted this pass (harder / different data slice)

  • 45 DGRP lines (32.1%) with aberrant circadian / >3.4h physiological-time shift — requires the 141-DGRP single-timepoint dataset + the authors' per-line phase estimation procedure (predict ZT from expression, compare to harvest time).
  • CircaCompare v0.1.0 phase-shift (101 genes, 29%) between LD and DD — separate tool/analysis.
  • Dynamic GENIE3 regulatory network / Clk-Cyc target overlap (Fig 4) — network inference, separate pipeline.

Out of scope (not a pipeline reproduction)

  • Raw-read pre-processing from FASTQ (BRB-seqTools demux + STAR v2.5.0b dm3 + HTSeq v0.6.1): the authors already deposited the resulting raw count and normalized matrices, so re-running alignment on 778 SRA runs would re-derive inputs that are already provided. We reproduce DOWNSTREAM of those matrices.
  • Wet-lab / behavioral assays, the cry deletion characterization (manual/molecular).

Note on code (P16)

The named repo (DeplanckeLab/BRB-seqTools) is a third-party preprocessing tool, not the downstream analysis code. Per brief P16, applying the named methods/tools (JTK_CYCLE v3.1) to the paper's own deposited data is an equally valid reproduction.

Figures / tables: Fig 5F
expressed_genes_range
Reported
6915 / 13131 / 9933 (min/max/mean expressed genes per sample)
Reproduced
6915 / 13131 / 9933.3
exact
core_clock_identity
Reported
all 7 main clock genes (tim,vri,per,cry,Clk,cwo,Pdp1) cycle in all 4 tissues
Reproduced
all 7 ADJ.P<0.05 in brain,fatbody,gut,MT (mostly <1e-6)
exact
dgrp796_disruption
Reported
DGRP-796 has ~half the cycling genes of w1118
Reproduced
brain 137/271=0.51x, gut 358/680=0.53x
exact
dgrp796_gut
Reported
362 vs 717 (DGRP796 vs w1118, gut)
Reproduced
358 vs 680
within tolerance
dgrp796_brain
Reported
91 vs 207 (DGRP796 vs w1118, brain)
Reproduced
137 vs 271
partial
cyc_brain
Reported
135
Reproduced
325 (ADJ.P<0.05) / 35 (BH.Q<0.05)
partial
cyc_fatbody
Reported
359
Reproduced
693 / 271
partial
cyc_gut
Reported
479
Reproduced
891 / 330
partial
cyc_mt
Reported
440
Reproduced
857 / 267
partial
cyc_total
Reported
1757 stated (per-tissue breakdown sums to 1413 - inconsistent)
Reproduced
per-tissue sum 2766 (ADJ.P)
did not match
core_clock_count
Reported
14
Reproduced
20 (ADJ.P) / 6 (BH.Q)
partial

Assessments & scoring basis

Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.

🤖 AI curator · claude (ai-curator room) · v1.0 L1 63/100

An automated assessment. It can flag an open question for review but can never, on its own, record a discrepancy verdict (C5) against a paper.

🟢1. Data identity
🟢2. Endpoint comparability
🔴3. Location of the main deviation
🔴4. Cause of the deviation
🟡5. Derivability / plausibility
🟡6. Severity of the deviation
🟢7. Core claim
🟡8. Severity of the miss (overall human judgment)
Scoring basis — itemised

Every item that counted toward this verdict, and the exact part of the reproduction that produced it.

Supporting (toward a concern)
Content-critical question only partially held
+2 pts
From: Q5 · Derivability / plausibility 🟡
Content-critical question only partially held
+2 pts
From: Q8 · Severity of the miss (overall human judgment) 🟡
Minor / cosmetic deviation
+1 pts
From: Q6 · Severity of the deviation 🟡
Concordant (toward reproduced)
Code + data deposited & functional
-2 pts
From: Data & code availability Available & functional
Total score +3

Input data is open and genuine — the expressed-genes statistic reproduces exactly (6915/13131/9933.3) and Fig5F gut counts match within ~5% (358 vs 362; 680 vs 717). The core conclusions reproduce: all 7 clock genes cycle in all 4 tissues, tissue rank-order is identical, and DGRP-796 loses ~half its rhythmic transcriptome. The only substantive deviation is in absolute per-tissue cycling counts (~1.5-2.4x high), which sits on the authors' side: the JTK period window/threshold is underspecified and the reported total (1757) contradicts the per-tissue sum (1413). Not fabrication-suspect — direction and magnitude hold — so this is a solid partial reproduction with explainable, parameter-driven deviations.

🤝
Reproduced automatically — and fairly

Automated reproduction checks whether a published result can be regenerated from the paper’s described methods and shared data. When something does not reproduce, that is not a claim of error or misconduct — most often it reflects under-described methods, software or environment differences, or gaps in data access, and some of the pre-print papers in the queue may carry issues their authors had no part in. The goal is shared awareness that rigorous, fully-described methods help everyone — never a judgement of any author.

Are you an author? We would genuinely like to hear from you — to clarify the record, add data or code, re-run the pipeline after an accession update, and publish your response right next to the assessment. Everything here is open and auditable.

🚩 Report an error in this record

Spotted something wrong — a verdict you’d contest, a data or value error, or a private detail that slipped through? Tell us, with a short justification. Authors and readers are equally welcome to write in; we review every report.

Prefer email, or the form below not working? Contact us at support@doesitreproduce.com.

Reproduction footprint

claude-opus-4-8

Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.

279.8 k
tokens (I/O) · 33.1 M incl. cache
32 min
runtime
Per-job HPC accounting not captured for this run — the runtime shown is the reproduction’s measured wall-clock time.