{"id":"c41c2081-f80a-4813-ab32-6760ba705492","arxiv_id":"2505.00572","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"Database mining plus a reanalysis of serum miRNA sequencing from MMA fighters yields 11 candidate genes and 8 candidate miRNAs, with hsa-miR-10a-5p highlighted as an acute mTBI biomarker candidate.","lead":"This in silico study screens gene and miRNA databases and reanalyzes a public RNA-seq dataset of mixed martial arts fighters to nominate 11 genes and 8 microRNAs, especially hsa-miR-10a-5p, as candidate mild traumatic brain injury biomarkers. A generalist might read it as a low-cost, hypothesis-generating blueprint for blood-based concussion diagnostics, though it stops short of clinical validation.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central biomarker claim depends on attributing post-fight serum miRNA changes to mTBI, but the GSE123336 design lacks a non-head-injury control; exercise or stress alone could drive hsa-miR-10a-5p, and the Discussion's own limitation admits this.","rationale":"The reader's weakest assumption and my load-bearing concern coincide: the absence of a non-head-injury control makes it impossible to attribute the serum miRNA changes to mTBI rather than exercise, stress, or extracranial trauma. This is not an external nitpick; the manuscript's own Discussion explicitly identifies the missing control group and states that such controls 'would help determine whether extracranial injuries impact the levels of candidate miRNAs.' Thus the central claim, that hsa-miR-10a-5p may serve as an mTBI biomarker, is conditional on specificity that the current dataset cannot establish. I still regard the analysis as a plausible hypothesis-generating screen: the authors acknowledge the limitation, and the result could be true. The reader's CONDITIONAL verdict already reflects this, so no verdict adjustment is needed. The concrete test would strengthen or falsify the specificity assumption by reusing the same pipeline on a dataset with exercise or non-head trauma controls.","tokens_in":23158,"tokens_out":3278,"duration_ms":34233,"concrete_test":"Re-run the same GSE123336 pipeline (BWA-hg38/miRBase alignment + DESeq2) on a publicly available acute-exercise or orthopedic-trauma miRNA dataset (e.g., serum miRNA profiles before/after exhaustive exercise or non-head-injury trauma) and compare effect size and direction for hsa-miR-10a-5p, hsa-miR-10b-5p, and hsa-miR-143-3p. If hsa-miR-10a-5p shows a comparable post-exercise elevation without head impact, the mTBI-specific interpretation fails. A secondary check: within GSE123336, test whether fight duration or other exertion proxies are associated with hsa-miR-10a-5p levels independent of head hits.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is the inference that post-fight serum miRNA elevations, especially hsa-miR-10a-5p, are caused by mTBI. In the RNA-seq analysis, 0-day post-injury samples are compared with each fighter's pre-injury baseline (Methods; GSE123336). Under this design, an mTBI is perfectly confounded with an acute bout of exhaustive exercise, psychological stress, and any systemic physical trauma from the bout. The original MMA dataset has no non-head-injury or sham-exercise control arm, and the authors explicitly concede in the Discussion: 'We were unable to include a control group for circulating biomarkers... Introducing controls from patients with extracranial injuries... would help determine whether extracranial injuries impact the levels of candidate miRNAs, which could affect their reliability as diagnostic tools for mTBI.' If hsa-miR-10a-5p responds to exertion or stress alone, the central claim that it is an mTBI biomarker collapses. The head-hit correlation for miR-10b-5p/miR-143-3p does not rescue this, because number of head hits is also correlated with fight intensity and physical load.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This in silico study aims to identify genes and miRNAs associated with mild traumatic brain injury (mTBI) severity and recovery. The authors compile candidate genes from NCBI Gene, MalaCards, and their own prior literature review, reduce the list to 11 hub genes through enrichment and protein-protein interaction analyses, predict eight putatively regulatory miRNAs using several target-prediction tools, and then analyze the public RNA-seq dataset GSE123336 (serum miRNAs from MMA fighters) to look for differential expression after head injury. They report that 17 miRNAs are significantly changed at day 0 and again at 2–3 days post-injury, highlight hsa-miR-10a-5p as particularly relevant, and note that hsa-miR-10b-5p and hsa-miR-143-3p correlate with the number of head hits. The abstract concludes that these genes and miRNAs, particularly hsa-miR-10a-5p, may guide future mTBI diagnostics.","tokens_in":23386,"tokens_out":4741,"duration_ms":49657,"significance":"If the central claim held, the proposed 11-gene/8-miRNA network and the specific serum miRNA hsa-miR-10a-5p would be plausible priority candidates for prospective mTBI biomarker validation. The study has some strengths: it uses publicly available high-throughput data, applies explicit significance thresholds, reports exact p-values, and transparently acknowledges the absence of a control group for circulating biomarkers. However, the current evidence is hypothesis-generating rather than confirmatory. The main analytical pipeline is largely reproducible in principle, although the authors state that code and analysis details are only available on request, and the overlap between the predicted miRNA list and the RNA-seq results is very small. The potential clinical significance is real but conditional on overcoming the confounding and circularity concerns detailed below.","major_comments":[{"comment":"The central inference that hsa-miR-10a-5p is an mTBI biomarker is confounded by the design of GSE123336. In the RNA-seq analysis, 0-day post-injury samples are compared with each fighter's pre-injury baseline (Methods, 'RNA-sequencing Analysis'), but an mTBI is perfectly confounded with the acute physiological stress of an MMA bout, including exhaustive exercise, psychological stress, and possible extracranial trauma. The authors themselves acknowledge in the Discussion that no control group was included and that extracranial injury controls would be needed to establish specificity. Given this acknowledged limitation, the abstract's statement that 'specific genes and miRNAs, particularly hsa-miR-10a-5p, may influence mTBI outcomes' and the framing of hsa-miR-10a-5p as a diagnostic candidate overstate what this design can support. The authors should either provide external validation with a non-head-injury control or explicitly reframe the conclusion as a hypothesis-generation result rather than a biomarker claim.","section":"Results, 'Identification of Differential Expression miRNAs in GSE123336'; Discussion, Limitations"},{"comment":"The reduction from 129 genes to 11 candidate genes is circular with respect to the reported enrichment results. Genes were selected using inclusion criteria that explicitly require 'functional similarity to established mTBI-related pathways' and involvement in 'neuroinflammation, neuronal repair, and synaptic plasticity' (Methods), and genes expressed only in brain tissue, inflammatory genes, and cytokines were then excluded post hoc. It is therefore not surprising that the final 11 genes are enriched for neuron projection regeneration and synaptic plasticity; the enrichment is partly guaranteed by the selection procedure. Furthermore, Table 1's top biological process, neuron projection regeneration, is based on only 2 of the 11 genes (count = 2) with a p-value of 4.80E-03 that is not corrected for the number of GO terms tested. The authors should present an unbiased gene-selection procedure, use a background-matched enrichment test, or explicitly state that the enrichment is a descriptive property of a curated list rather than independent evidence.","section":"Methods, 'Data Analysis: Functional Enrichment and Pathway Analysis'; Results, 'Identifying Candidate Genes Using…"},{"comment":"The claimed validation of predicted miRNAs by RNA-seq rests on a single overlapping miRNA. Of the eight predicted miRNAs listed in Table 3, only hsa-miR-10a-5p appears among the 17 differentially expressed miRNAs in Table 4. The two miRNAs highlighted for their correlation with head hits, hsa-miR-10b-5p and hsa-miR-143-3p, are not in the predicted miRNA set. The manuscript does not provide an intersection analysis, an expected-overlap calculation, or any adjustment for testing the overlap between an 8-miRNA prediction set and a 17-miRNA DE set. As a result, the 'comparison with our miRNA predictions' is at most a single concordant observation, and the text should quantify this overlap and temper the claim that the RNA-seq analysis validates the predicted miRNA panel.","section":"Results, 'miRNA-Target Analysis'; Table 3; Table 4"},{"comment":"The statistical reporting is internally inconsistent and incomplete. The Methods state that a P-adjusted value threshold of ≤ 0.05 was used and then say a Bonferroni correction was applied, but DESeq2's default 'padj' is the Benjamini-Hochberg FDR, not a Bonferroni correction. If a custom Bonferroni correction was applied, the parameters are not described and the code is not deposited. In addition, the text says 17 miRNAs were significantly changed at both 0 days and 2–3 days post-injury, but Table 4 lists results only for the 0-day comparison, and no table or supplementary file gives the 2–3 day p-values and directions of change. The authors should clarify which multiple-testing correction was actually used and provide the full 2–3 day results.","section":"Methods, 'Data Processing'; Results, 'Identification of Differential Expression miRNAs in GSE123336'; Table 4"},{"comment":"The claim that hsa-miR-10b-5p and hsa-miR-143-3p 'were positively correlated with direct head hits' is not supported by any reported correlation coefficient, confidence interval, or statistical test. The text provides only adjusted p-values for a comparison between participants with more than 20 head hits and other participants. A comparison of groups is not itself a correlation, and the number of head hits is also likely to correlate with fight intensity, physical exertion, and total trauma burden. The authors should report the actual correlation analysis or reframe the statement as a group comparison, and they should acknowledge that this association is not specific to head trauma.","section":"Results, 'Identification of Differential Expression miRNAs in GSE123336'"}],"minor_comments":[{"comment":"Several miRNA names are misspelled as 'has-miR-218-5p', 'has-miR-34a-5p', and 'has-miR-199b-5p'; these should be corrected to 'hsa-miR-' throughout the manuscript.","section":"Abstract; Results, 'miRNA-Target Analysis'"},{"comment":"The statement 'The coding and analysis details are available upon request' (Data Availability) is not sufficient for reproducibility; the authors should deposit the analysis scripts and parameter settings in a public repository.","section":"Methods, 'Data Processing'"},{"comment":"The table headers contain spelling errors: 'GOBP_SYNSPTIC_SIGNALING' should be 'GOBP_SYNAPTIC_SIGNALING', and 'HP_COGNITTIVE_IMPAIRMENT' should be 'HP_COGNITIVE_IMPAIRMENT'.","section":"Table 2"},{"comment":"The sentence ending '...hsa-miR-143-3p (Padj =0.0082) (Figure 7In terms of GO and pathway analysis...' is missing a period and a space; it should read 'Figure 7. In terms of GO and pathway analysis...'.","section":"Results, 'Identification of Differential Expression miRNAs in GSE123336'"},{"comment":"The sentence beginning 'Lastly, for future studies, integrating bulk RNA-seq with single-cell transcriptomics...' introduces DLK inhibition and ATF3 modulation without a connecting explanation; the relevance to the preceding miRNA discussion should be clarified or the sentence shortened.","section":"Discussion"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a hypothesis-generating bioinformatics screen with acknowledged limitations. Its central biomarker claim is not supported as stated because of the confounding in GSE123336 and the very small overlap between predicted and observed miRNAs. These issues are fixable within the manuscript's scope by re-framing the claims, adding appropriate statistical analyses, and providing the missing results. I do not see evidence of fabrication or misconduct; the main concern is over-interpretation relative to the evidence presented."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: this is a competent, hypothesis-generating bioinformatics screen. The new bit is the specific combination: 11 curated genes, 8 predicted regulatory miRNAs, and a re-analysis of the MMA serum miRNA dataset with head-hit counts. The paper frames the results as candidates for validation, not as established biomarkers, which is the right framing.\n\nWhat it does well: the workflow is standard but executed sensibly. Using multiple miRNA prediction tools and requiring overlap is reasonable. The RNA-seq re-analysis of GSE123336 with DESeq2 and Bonferroni is defensible as a screen. The Discussion is refreshingly explicit about the biggest weakness: no control group for circulating biomarkers, and sample numbers drop sharply at later timepoints. They also acknowledge that excluding brain-specific genes could limit the findings.\n\nThe soft spots are real, and one is load-bearing. Serum miRNA changes in MMA fighters are measured after a fight, relative to each fighter's own baseline, with no non-head-injury or sham-exercise arm. Exercise, stress, and systemic trauma from the bout are all confounded with mTBI. The authors admit this in the Discussion, but it means hsa-miR-10a-5p, the headline finding, cannot be attributed to mTBI from this dataset. The head-hit correlation for miR-10b-5p and miR-143-3p is a bit stronger, but number of head hits also tracks fight duration and intensity, so it is still not a clean mTBI-specific signal.\n\nOther soft spots in proportion: the reduction from 129 genes to 11 is subjective and post hoc; no code or detailed parameters were released; and the \"correlation\" with head hits is actually a comparison of fighters with >20 hits versus fewer, not a computed correlation. Minor: the abstract says \"2664 differentially expressed miRNAs,\" which is misleading; that is the number of miRNAs analyzed, not the DE hits. The citation pattern is fine; the self-citation is to their own prior review, which is legitimate.\n\nWho is this for? People working on mTBI miRNA biomarker discovery who want a prioritized list to test in prospective cohorts. It deserves a serious referee, but I would not accept any causal claim from the RNA-seq part.\n\nI'd send it to peer review as a bioinformatics screen, conditional on releasing code and parameters and toning down the causal language in the abstract. The candidate list is useful; the miRNA specificity claim is not yet supported.","headline":"A competent hypothesis-generating bioinformatics screen whose central miRNA claim is undercut by the uncontrolled MMA dataset; worth a serious referee but only as a candidate list.","tokens_in":23924,"tokens_out":2125,"would_cite":false,"duration_ms":21781,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Circulating microRNA hsa-miR-10a-5p rises after mild head injury and may serve as a blood-based biomarker of severity and recovery.","keywords":["mild traumatic brain injury","microRNA","biomarker","hsa-miR-10a-5p","hub genes","RNA sequencing analysis","bioinformatics","synaptic plasticity"],"falsifier":"Collect serum from a cohort with extracranial injuries (for example orthopedic trauma) and from athletes after intense exercise without head impacts, then measure hsa-miR-10a-5p, hsa-miR-10b-5p, and hsa-miR-143-3p. If these miRNAs rise in either control group to levels seen after mixed martial arts bouts, the mTBI-specific biomarker claim fails.","tokens_in":22950,"feed_emoji":"🧠","tokens_out":5826,"duration_ms":53867,"temperature":0.7,"pith_summary":"The paper argues that a small set of genes and microRNAs, led by hsa-miR-10a-5p, tracks the severity and recovery of mild traumatic brain injury. Using public databases and a serum RNA-sequencing dataset from amateur mixed martial arts fighters, it identifies 11 hub genes engaged in neuron projection regeneration and synaptic plasticity, plus 8 miRNAs predicted to regulate them. It then finds 17 miRNAs significantly changed immediately after injury, with hsa-miR-10a-5p strongly upregulated, and links hsa-miR-10b-5p and hsa-miR-143-3p to the number of head hits. If correct, these molecules would be priority candidates for non-invasive diagnostic and prognostic tests for mTBI.","feed_headline":"Blood microRNA hsa-miR-10a-5p jumps after mild head injury","feed_subtitle":"A bioinformatic screen singles out 11 genes and 8 miRNAs that could diagnose mTBI and track recovery.","key_machinery":"The load-bearing object is the predicted regulatory network linking the 11 hub genes to eight miRNAs: hsa-miR-9-5p, hsa-miR-204-5p, hsa-miR-1908-5p, hsa-miR-16-5p, hsa-miR-10a-5p, hsa-miR-218-5p, hsa-miR-34a-5p, and hsa-miR-199b-5p. The hub genes were selected from 129 candidates by functional enrichment, protein-protein interaction topology, and exclusion of brain-only and inflammatory genes; the miRNAs were kept only if multiple prediction tools agreed. This network then anchors the interpretation of differential expression in serum after acute injury, connecting observed miRNA rises to biological processes of neuronal repair.","core_discovery":"The paper's central claim is that specific genes and miRNAs, particularly hsa-miR-10a-5p, influence mild traumatic brain injury outcomes and that measuring them in blood could help diagnose injury and forecast recovery. The supporting argument is built from an 11-gene hub set: APOE, S100B, GFAP, BDNF, AQP4, COMT, MBP, UCHL1, DRD2, ASIC1, and CACNA1A, whose functional enrichment converges on neuron projection regeneration and synaptic plasticity. Eight miRNAs predicted to target these genes were identified, and in the serum dataset 17 miRNAs changed significantly at 0 and 2-3 days post-injury, with hsa-miR-10a-5p among the most strongly elevated immediately after injury. The paper also reports that hsa-miR-10b-5p and hsa-miR-143-3p levels correlate positively with the number of direct head hits, making them candidate markers of injury load.","pith_inferences":["A prospective study adding an orthopedic-trauma control group would directly test whether hsa-miR-10a-5p is brain-specific rather than a general tissue-damage marker; such a design is the natural next step.","The dopaminergic synapse enrichment suggests the same miRNA set may be altered in non-traumatic conditions such as depression or Parkinson's disease, so cross-disorder serum profiling would clarify what is mTBI-specific.","Because the fighter dataset cannot separate concussion from cumulative subconcussive hits, an experiment measuring these miRNAs after controlled soccer heading without diagnosed concussion would test whether subconcussive impacts alone trigger the same rises.","Single-cell transcriptomic data from injured cortex could reveal whether hsa-miR-10a-5p and BDNF are co-expressed in regenerating projection neurons, giving mechanistic support to the predicted regulatory link."],"forward_implications":["Blood levels of the 17 acutely changed miRNAs, especially hsa-miR-10a-5p, could become a diagnostic readout for recent mild head injury.","The relationship of hsa-miR-10b-5p and hsa-miR-143-3p with number of head hits could provide a molecular measure of mechanical injury load.","The 11-gene network centered on neuron projection regeneration and synaptic plasticity could nominate repair pathways for therapeutic targeting.","The apparent normalization of miRNA levels by one week after injury implies the blood biomarker window is acute, limiting its use for retrospective diagnosis."],"supporting_citations":[{"why":"Supplies the serum RNA-sequencing dataset from mixed martial arts fighters; source of the differential-expression findings.","marker":"[45]"},{"why":"The authors' earlier literature review; provides the initial 30-gene candidate pool and inclusion rationale.","marker":"[46]"},{"why":"The statistical method used to call differentially expressed miRNAs from the sequencing counts.","marker":"[73]"},{"why":"Prior evidence that miRNAs distinguish mild from severe TBI; the comparison point for the paper's claim that miRNAs may outperform protein biomarkers.","marker":"[43]"},{"why":"Provides the mature miRNA sequence annotation used to align and count sequencing reads.","marker":"[71]"},{"why":"One of the target-prediction tools whose consensus filtered the eight miRNAs linked to the hub genes.","marker":"[57]"}],"fun_headline_variants":["Blood miR-10a-5p surges right after mild brain injury","MicroRNA panel predicts mild TBI recovery outcome","One microRNA, 11 genes flag mTBI severity","In silico study finds microRNA biomarker for mTBI","hsa-miR-10a-5p: early blood marker for concussion"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The serum miRNA changes measured in mixed martial artists after a bout are due to mild brain injury rather than to exercise, physiological stress, or extracranial trauma; the dataset had no non-head-injury control group.","fun_headline_variants_meta":{"raw":{"variants":["Blood miR-10a-5p surges right after mild brain injury","MicroRNA panel predicts mild TBI recovery outcome","One microRNA, 11 genes flag mTBI severity","In silico study finds microRNA biomarker for mTBI","hsa-miR-10a-5p: early blood marker for concussion"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000456,"raw_usage":{"total_tokens":2392,"prompt_tokens":1153,"completion_tokens":1239,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":769,"completion_tokens_details":{"reasoning_tokens":1151}},"tokens_in":769,"tokens_out":1239,"duration_ms":11850,"temperature":1.0,"reasoning_tokens":1151,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T04:39:13.996855+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Collect serum from a cohort with extracranial injuries (for example orthopedic trauma) and from athletes after intense exercise without head impacts, then measure hsa-miR-10a-5p, hsa-miR-10b-5p, and hsa-miR-143-3p. If these miRNAs rise in either control group to levels seen after mixed martial arts bouts, the mTBI-specific biomarker claim fails.","supporting_citations":[{"cited_title":"Comparison of serum and saliva miRNAs for identification and characterization of mTBI in adult mixed martial arts fighters","cited_arxiv_id":null,"evidence_quote":"Supplies the serum RNA-sequencing dataset from mixed martial arts fighters; source of the differential-expression findings."},{"cited_title":"Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2","cited_arxiv_id":null,"evidence_quote":"The statistical method used to call differentially expressed miRNAs from the sequencing counts."},{"cited_title":"miRBase: from microRNA sequences to function","cited_arxiv_id":null,"evidence_quote":"Provides the mature miRNA sequence annotation used to align and count sequencing reads."},{"cited_title":"miRSystem: An Integrated System for Characterizing Enriched Functions and Pathways of MicroRNA Targets","cited_arxiv_id":null,"evidence_quote":"One of the target-prediction tools whose consensus filtered the eight miRNAs linked to the hub genes."}],"review_version":1}