Plexin B3 promotes neurite outgrowth, interacts homophilically, and interacts with Rin.
Part of the results reproduced; minor but material deviations remained.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
- ✓No authors-side cause for any deviation
- ✓Any deviation was negligible
- 🔴Could not use the authors’ exact input data
- 🔴Reported values were only indirectly comparable
- 🟡A deviation arose in the data or preprocessing
- 🟡Reported values were not (fully) derivable from the shared data
- 🟡The central claim did not (fully) hold under reproduction
- 🟡Overall, the reproduction showed a material discrepancy
▸Reproduction agent’s raw note
PARTIAL reproduction. 'Plexin B3 promotes neurite outgrowth, interacts homophilically, and interacts with Rin' (Hartwig et al., BMC Neuroscience 2005;6:53; PMC1215486) is predominantly a wet-lab study (neurite-outgrowth microscopy, cell-aggregation assays, yeast two-hybrid, co-IP, Western, biotinylation, in situ) with NO deposited experimental dataset (no GEO/SRA/ENA/PRIDE) and NO analysis code. The one genuinely reproducible COMPUTATIONAL result is the conceptual translation of the open reference cDNA AF149019 (human PLXNB3), reported as a 1909-aa, 207-kDa, pI-6.21 protein. Reproduced inside SLURM «job» on «our HPC» compute node n094 (Biopython 1.87, Bio.Entrez/SeqIO + ProtParam): protein length 1909 aa = EXACT; molecular weight 206.84 kDa rounds to 207 kDa = EXACT on the paper's rounding; isoelectric point 5.96 vs reported 6.21 = WITHIN-TOL (delta 0.25, the well-known pKa-scale difference between Biopython ProtParam and the authors' ExPASy/Bjellqvist Compute pI/Mw). The dbEST BLAST ('56 matching entries') is NOT reproducible because NCBI retired dbEST in 2019 and the match criterion is undefined; the neurite ANOVA (p<=0.0033) is NOT recomputable because per-neuron microscopy data was never deposited. No fabrication concern on the translation values — all are directly derivable from the open reference cDNA. Honest outcome: one computational claim reproduced (length exact, MW exact, pI within tolerance); wet-lab majority out of scope.
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.
-
v1 current initial assessmentassessed: 2026-06-19 ⛓ cf71742d448e
✎ 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.
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
- 2026-06-22
- Rubric version
- v1.0
- Assessed by
-
🤖 AI curator · claude (ai-curator room) · v1.0 · run #1 2026-06-19no 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: sonnetThe paper investigates the subtype-specific functions of B-family plexins (B2 and B3) beyond their known role as semaphorin receptors in axon repulsion, testing whether B3 (and B2) can positively influence neuronal morphogenesis through homophilic interaction and intracellular signaling partners.
- ★ Plexin B3 strongly and plexin B2 moderately stimulate neurite outgrowth of primary murine cerebellar neurons finding
- ★ B3 and B2 mediate Ca2+/Mg2+-dependent homophilic trans-interaction causing cell aggregation finding
- ★ The sema domain of B3 is essential and sufficient for homophilic interaction mechanism
- ★ The neuron-specific, calmodulin-binding Ras-related GTPase Rin interacts with the intracellular part of B3 but not B2 finding
- ★ Rin co-localizes and co-immunoprecipitates with B3 in co-transfected COS-7 cells finding
- PLXNB3 is expressed abundantly in brain, with alternative splicing of exon 27 generating at least three isoforms finding
- B3 protein undergoes proteolytic processing and N-glycosylation, yielding a mature ~260 kDa cell-surface form finding
- B3 shows no heterophilic co-immunoprecipitation interaction with plexins A1, B1, or B2 finding
| Assay | System | Perturbation | Readout | Platform |
|---|---|---|---|---|
| Northern blot | 12 human adult tissues | none | PLXNB3 mRNA expression profile | poly(A)+ mRNA northern blot |
| Western blot | COS-7 cells overexpressing human B3 / human brain (neocortex, corpus callosum) / mouse brain and COS-7 cells expressing mouse B3 | overexpression | B3 protein detection, size, and processing | antibodies pAbB3-A, pAbB3-B, pAbmB3 |
| In situ hybridization / immunocytochemistry / immunohistochemistry | primary murine cerebellar neurons, astrocytes, oligodendrocytes; adult mouse cerebellum; adult human cerebellum | none | PlxnB3 mRNA and B3 protein localization | — |
| Neurite outgrowth assay | primary murine cerebellar neurons (6-day-old C57BL/6J mice) grown on NIH-3T3 substrate cells expressing L1, B2, or B3 | overexpression of L1/B2/B3 in substrate cells | neurite length | light microscopy |
| Cell aggregation assay (DiI/DiO labeling) | NIH-3T3 cells stably expressing B3, B2, or L1 | overexpression; presence/absence of Ca2+/Mg2+ | aggregation index (particle number over time) | fluorescence microscopy |
| Co-immunoprecipitation | COS-7 cells co-transfected with full-length and deletion constructs of B3 (and plexins A1, B1, B2) | deletion constructs / co-expression | homophilic and heterophilic interaction domain mapping | Western blot |
| Yeast two-hybrid screen | yeast, screened against intracellular domain of B3 | none | identification of Rin as B3-interacting protein | yeast two-hybrid |
| Co-immunoprecipitation / co-localization | COS-7 cells co-transfected with B3 and Rin | co-expression | B3-Rin protein interaction | Western blot / immunocytochemistry |
- ▲ B3-expressing substrate cells produced the strongest neurite outgrowth stimulation (mean 129 ± 58 μm) versus nontransfected control (51 ± 30 μm) ~2.5-fold
- ▲ B2-expressing substrate cells stimulated neurite outgrowth (mean 84 ± 37 μm) versus nontransfected control ~1.6-fold
- – Mean neurite length differed significantly across all substrate groups ANOVA p ≤ 0.0033
- ▲ After 80 min, aggregation-driven particle number decrease was greatest for L1 (79%) and B3 (76%) expressing cells, less for B2 (64%), least for nontransfected cells (52%) 76% decrease (B3)
- ▼ Removal of divalent cations abolished B2- and B3-dependent aggregation but not L1-dependent aggregation
- – Sema domain alone is necessary and sufficient for B3 homophilic co-immunoprecipitation, while constructs lacking the sema domain fail to interact
- – No heterophilic co-immunoprecipitation interaction detected between B3 and plexins A1, B1, or B2
- – Rin co-localizes and co-immunoprecipitates with B3 in co-transfected COS-7 cells
- pvalue p ≤ 0.0033 (ANOVA comparing neurite lengths across substrate cell groups, minimum 400 neurons analyzed per group)
- mean 51 μm ± 30 (S.D.) (mean neurite length on nontransfected NIH-3T3 control substrate)
- mean 84 μm ± 37 (S.D.) (mean neurite length on B2-expressing substrate cells)
- mean 93 μm ± 59 (S.D.) (mean neurite length on L1-expressing substrate cells)
- mean 129 μm ± 58 (S.D.) (mean neurite length on B3-expressing substrate cells)
- other particle number decreased 79%, 76%, 64%, 52% for L1-, B3-, B2-, and non-transfected cells respectively (cell aggregation assay after 80 min)
- other 1,909 aa, 207 kDa, pI 6.21 (predicted properties of full-length human B3 protein from cDNA translation)
- count 56 matching EST entries (BLASTn screening of human dbEST using AF149019 sequence)
Statistical methods review
Model: sonnetA neutral, descriptive read of the statistical approach — what was done, and (for shared learning, not as criticism) what could also have been done.
The paper reports a series of cell-biology and biochemical experiments (Western blot, co-immunoprecipitation, in situ hybridization, cell aggregation assays, and neurite outgrowth assays) comparing plexin B2- and B3-expressing substrate cells to L1-expressing and non-transfected controls. The main quantitative comparison, neurite outgrowth length across four substrate groups, was tested by one-way ANOVA with results reported as mean ± S.D. or mean with S.E.M. error bars from pooled data across independent experiments. Most other findings (Western blots, co-IP, ISH, aggregation assay time courses) are described and shown qualitatively/graphically rather than with formal inferential statistics.
| Test | Applied to | n | Assumptions |
|---|---|---|---|
| one-way ANOVA | comparison of mean neurite length across four substrate groups (non-transfected 3T3, B2, L1, B3) in Figure 6B | a minimum of 400 neurons analyzed per substrate cell type, pooled from three independent experiments | not stated |
-
Overall group differences in neurite outgrowth were assessed with a one-way ANOVA across four substrate groups, with pairwise significance markers (*, **) shown in the figure without a named post-hoc test.↳ Could also: A named post-hoc procedure such as Tukey's HSD or Dunnett's test (comparing each group to the non-transfected control) could also be applied following the ANOVA — Naming and applying a specific post-hoc method makes explicit how the pairwise p-values were derived and how the family-wise error rate across the multiple pairwise comparisons was controlled.
-
Neurite length data were pooled across three independent experiments before ANOVA, with substrate type as the grouping factor.↳ Could also: A linear mixed-effects model with experiment (biological/technical replicate) as a random effect and substrate type as a fixed effect could also be used — Explicitly modeling experiment-to-experiment variation as a random effect can account for potential clustering or batch effects within each of the three independent experiments, complementing the pooled-data ANOVA.
-
Dispersion is reported as S.D. for the neurite length summary in the text and as S.E.M. for the corresponding figure and aggregation assay figures.↳ Could also: Reporting a 95% confidence interval for the group means, or using S.D. consistently throughout, could also be used — A confidence interval directly conveys the precision of the estimated mean and its plausible range, and using one consistent dispersion measure throughout the figures and text can make effect magnitudes easier to compare across analyses.
-
Statistical significance for neurite outgrowth differences is reported via ANOVA p-values without an accompanying effect-size statistic.↳ Could also: An effect-size measure such as eta-squared (for the ANOVA) or Cohen's d (for pairwise group comparisons) could also be reported — Effect sizes quantify the magnitude of the difference in neurite outgrowth between groups independent of sample size, complementing the significance testing already reported.
-
No explicit statement is given on whether neurite length data met ANOVA's normality and homogeneity-of-variance assumptions.↳ Could also: A non-parametric alternative such as the Kruskal-Wallis test (with Dunn's post-hoc test for pairwise comparisons) could also be used — A non-parametric approach can be applied without relying on distributional assumptions about neurite length, which may be useful given the right-skewed appearance typical of neurite length distributions.
-
Aggregation assay results (Figure 10) are presented as mean ± S.E.M. across three independent experiments without a formal statistical test comparing aggregation kinetics between cell types.↳ Could also: A repeated-measures ANOVA or a mixed model comparing aggregation index across time points and cell types could also be applied — Formal testing of the time-course data could quantify whether the differences in aggregation kinetics between B2-, B3-, L1-, and non-transfected cells are statistically distinguishable at each time point.
What was reproduced
The exact results taken into scope, with each reported value next to the value our attempt produced.
Scope — PMID 16122393
Title: Plexin B3 promotes neurite outgrowth, interacts homophilically, and interacts with Rin. Authors: Hartwig C, Veske A, Krejcova S, Rosenberger G, Finckh U. Journal / year: BMC Neuroscience 2005;6:53. DOI: 10.1186/1471-2202-6-53 · PMID: 16122393 · PMCID: PMC1215486 Full text reviewed: PubMed abstract + open-access full text (PMC1215486).
Verdict: PARTIAL — predominantly wet-lab; ONE computational result reproduced
This is a predominantly wet-lab molecular/cell-biology paper. There is no analysis code repository, no sequencing/array/proteomics data accession (no GEO/SRA/ENA/PRIDE/ArrayExpress), and no deposited experimental dataset. The vast majority of results are bench experiments (out of scope, not attempted).
Update (this run): the paper does report ONE genuinely reproducible
computational (sequence-analysis) result — the conceptual translation of the
open reference cDNA AF149019 (human PLXNB3): "a protein of 1,909 aa, molecular
mass 207 kDa, isoelectric point 6.21". This IS in scope (a deterministic
computation on an open reference sequence, reproducible with standard tools) and
was reproduced on «our HPC» (SLURM «job», Biopython 1.87): length 1909 aa
exact, MW 206.84 kDa → 207 kDa exact on rounding, pI 5.96 vs 6.21
(within-tol, pKa-scale dependent). See reproduction/agreement.json. The room is
therefore graded partial (one computational claim reproduced; wet-lab majority
out of scope; dbEST BLAST irreproducible), not a bare non_pipeline drop.
What the paper actually does (all wet-lab → not attempted)
- Neurite outgrowth assay — primary mouse cerebellar neurons on substrate; neurite length measured by microscopy; group means compared by ANOVA.
- Cell aggregation assay — fluorescent-labelled cell mixing; aggregation index Nt/N0 over time.
- Homophilic interaction / sema-domain mapping — cell aggregation + binding.
- Yeast two-hybrid screen — identifies Rin (neuron-specific GTPase) as an intracellular interaction partner of plexin B3.
- Co-immunoprecipitation, Western blot, cell-surface biotinylation, glycosylation analysis, in situ hybridization, immunocytochemistry, confocal microscopy.
All of the above are bench experiments quantified by hand/microscopy. None is a reproducible computational pipeline.
The only "computational" touches (still not a reproducible pipeline)
- One-off BLASTn note. Methods state: "BLASTn screening of human dbEST by
AF149019 revealed 56 fully matching entries." This is:
- run against dbEST, a database NCBI retired in 2019 (records folded into GenBank/nt); the ~2004 snapshot that produced "56 entries" no longer exists;
- undefined match criteria ("fully matching" — identity/coverage cutoff not given);
- therefore not reproducible 1:1 — a present-day BLAST would query a
differently-sized database and return a different count. It is a minor
methods remark, not a pipeline-derived result, and does not rescue the paper
from
non_pipeline.
- Manual domain annotation. sema domain + three cysteine-rich PSI/MRS repeats described from the literature/protein annotation — manual, no pipeline.
Reference sequences cited (NOT experimental datasets)
These are pre-existing reference/cDNA/EST GenBank entries the authors refer to, not data they deposited or that a pipeline consumes:
AF149019— human PLXNB3 full-length cDNA (reference).BF345653,H51489— human EST variants (reference).NM_019587— mouse plexin B3 mRNA (RefSeq reference). They are recorded indata/dataset_profile.jsonas reference accessions withaccess=open, but none constitutes a deposited experimental dataset the paper's results are computed from.
Drop reason
non_pipeline — text/topic looks bioinformatics-adjacent (plexins, sequences)
but the publication is a wet-lab study with no computational p
Assessments & scoring basis
Each contributor’s verdict, the per-question basis, and the auditable, itemised worksheet behind it.
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.
Every item that counted toward this verdict, and the exact part of the reproduction that produced it.
PMID 16122393 is a 2005 wet-lab paper (BMC Neuroscience, PMC1215486) reporting neurite-outgrowth microscopy, aggregation, yeast two-hybrid and co-IP — with no pipeline, no code, and no deposited dataset. This is a well-founded non_pipeline drop: q1/q2 are red (no comparable computational input/endpoint; raw per-neuron data not deposited), but q5/q7 stay yellow because the non-reproducibility is pure unavailability (uncheckable), not a defect or fabrication. The single computational touch (BLASTn vs dbEST = 56 ESTs) is irreproducible solely because NCBI retired dbEST in 2019, an external time-based loss rather than an authors' error.
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-4-8Measured resources invested to assess this paper — sanitised (machine class only, no job ids/paths). Compute = HPC accounting (SLURM); tokens = the AI agent's session.