Machine learning-based transcriptomic analysis identifies NAMPT and SAT1 as potential biomarkers and therapeutic targets in ferroptosis-associated rheumatoid arthritis.

{"sci_version": "1.2", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:41021591", "kind": "science-record", "revision": 4, "content_hash": "sha256:49acc1a5d86ee00c745994390a67810383947697f785291a944c2201b15ee671", "created_at": "2026-11-01T20:31:05Z", "updated_at": "2026-11-01T22:57:13Z", "generator": {"name": "corpus-assay-engine", "run_id": "reading-sci-2026-11-01"}, "paper": {"state": "complete", "title": "Machine learning-based transcriptomic analysis identifies NAMPT and SAT1 as potential biomarkers and therapeutic targets in ferroptosis-associated rheumatoid arthritis.", "doi": "10.1371/journal.pone.0333556", "pmid": "41021591", "authors": ["Devi Soorya Narayana Sasikumar", "Vino Sundararajan"], "journal": "PloS one", "year": 2025, "abstract": "BACKGROUND: Rheumatoid arthritis (RA) is an autoimmune disease with chronic presentation, involving symmetric joints and systemic involvement. Ferroptosis is iron-dependent programmed cell death through lipid peroxide accumulation, implicated in inflammatory diseases, including RA. However, its underlying mechanisms and gene-level contributions to RA pathogenesis remain largely unexplored. Therefore, this study emphasizes identifying ferroptosis-related genes associated with RA, evaluating their diagnostic, prognostic, and therapeutic potential, and exploring their role in immune modulation.\nMETHODS: The transcriptomic dataset (GSE89408) from the peripheral blood gene expression was downloaded from the Gene Expression Omnibus (GEO) database. We extracted the differentially expressed genes (DEGs) using R software and the most relevant modules relevant to RA were identified through weighted gene coexpression network analysis (WGCNA). We also identified the differentially expressed ferroptosis genes. The gene ontology and pathways involving the common genes were identified and the protein-protein interaction network was constructed. The hub genes were identified using three machine learning algorithms, least absolute shrinkage and selection operator (LASSO), random forest (RF), and support vector machine (SVM), after which the diagnostic efficiency of the hub genes and the correlation with immune infiltrating cells were predicted.\nRESULTS: A total of 9176 DEGs and a module of 314 genes were obtained which has a significant correlation with RA and 17 genes were selected after the intersection. Using the three machine-learning algorithms, we retrieved 8 hub genes (CISD2, LACTB, PRNP, SAT1, NAMPT, MITD1, SOD2, and FASN) between RA and ferroptosis which showed good diagnostic performance using the ROC curve and nomogram plots. Functional annotation analysis was utilized to inspect the biological functions of the hub genes and the genes showed a substantial association with the immune infiltrating cells.\nCONCLUSION: NAMPT, CISD2, LACTB, PRNP, SAT1, SOD2, MITD1, and FASN may modulate ferroptosis and RA by influencing immunity, and NAMPT and SAT1 contribute significantly to the diagnosis and treatment of the disease. Future studies focusing on validating these genes in larger cohorts and exploring their therapeutic potential will provide deeper insights.", "claims": [{"id": "claim-method", "statement": "We applied machine learning algorithms, including LASSO, random forest (RF), and support vector machine-recursive feature elimination (SVM-RFE), to identify the key genes, discern the hub genes, assess their diagnostic potential, and correlate them with immune cell infiltration patterns.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:41021591", "locator": "sections[1] \"1. Introduction\" paragraph 5", "quote": "We applied machine learning algorithms, including LASSO, random forest (RF), and support vector machine-recursive feature elimination (SVM-RFE), to identify the key genes, discern the hub genes, assess their diagnostic potential, and correlate them with immune cell infiltration patterns."}], "state": "complete"}, {"id": "claim-data-access", "statement": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21].", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:41021591", "locator": "sections[3] \"2.1. Data collection\" paragraph 2", "quote": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21]."}], "state": "complete"}, {"id": "claim-target-result", "statement": "Using the three machine-learning algorithms, we retrieved 8 hub genes (CISD2, LACTB, PRNP, SAT1, NAMPT, MITD1, SOD2, and FASN) between RA and ferroptosis which showed good diagnostic performance using the ROC curve and nomogram plots.", "kind": "qualitative", "target": null, "evidence": [{"asset": "source:pmid:41021591", "locator": "sections[0] \"Abstract\" paragraph 3", "quote": "Using the three machine-learning algorithms, we retrieved 8 hub genes (CISD2, LACTB, PRNP, SAT1, NAMPT, MITD1, SOD2, and FASN) between RA and ferroptosis which showed good diagnostic performance using the ROC curve and nomogram plots."}], "state": "complete"}], "gaps": []}, "experiment": {"state": "partial", "objective": "Using the three machine-learning algorithms, we retrieved 8 hub genes (CISD2, LACTB, PRNP, SAT1, NAMPT, MITD1, SOD2, and FASN) between RA and ferroptosis which showed good diagnostic performance using the ROC curve and nomogram plots.", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The claim, method and evidence span published with the source article.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:41021591"], "outputs": ["the published estimate and its evidence"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data from the named repository", "kind": "data-preparation", "state": "complete", "blocked_on": null, "description": "The GEO series accession resolves and its sample accessions are listed. This lane did not download the dataset or read its licence.", "duration_hours": null, "parameters": {}, "inputs": ["data:geo:GSE89408"], "outputs": ["the source data with its repository provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the published analysis", "kind": "analysis", "state": "blocked", "blocked_on": "method:machine_learning_random_forest", "description": "Apply the paper's statistical method to the resolved series and compare the result with the published target.", "duration_hours": null, "parameters": {}, "inputs": ["data:geo:GSE89408"], "outputs": ["the replicated result and its comparison"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:run_analysis blocked on method:machine_learning_random_forest"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:41021591", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.1371/journal.pone.0333556", "sha256": "sha256:637d41b4c7e2aec1a7847d13c330f1447d733e14ae036a31eaf0e18100f88278", "bytes": 52585, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.1371/journal.pone.0333556", "version": null, "note": "Full article fetched from PMC; the checksum covers the rendered section body, not the original XML bytes.", "structure": {"pmcid": "PMC12478934", "source": "europe_pmc_fulltext", "coverage": "whole_body_without_references", "text_hash_scope": "rendered_section_body"}}, {"id": "data:geo:GSE89408", "role": "raw", "format": "geo-series", "media_type": "application/vnd.ncbi.geo-series+json", "uri": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE89408", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE89408", "version": null, "note": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21].", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21].", "access_statement_kind": "repository_route_statement", "sample_accessions": ["GSM2370970", "GSM2370971", "GSM2370972", "GSM2370973", "GSM2370974", "GSM2370975", "GSM2370976", "GSM2370977", "GSM2370978", "GSM2370979", "GSM2370980", "GSM2370981", "GSM2370982", "GSM2370983", "GSM2370984", "GSM2370985", "GSM2370986", "GSM2370987", "GSM2370988", "GSM2370989", "GSM2370990", "GSM2370991", "GSM2370992", "GSM2370993", "GSM2370994", "GSM2370995", "GSM2370996", "GSM2370997", "GSM2370998", "GSM2370999", "GSM2371000", "GSM2371001", "GSM2371002", "GSM2371003", "GSM2371004", "GSM2371005", "GSM2371006", "GSM2371007", "GSM2371008", "GSM2371009", "GSM2371010", "GSM2371011", "GSM2371012", "GSM2371013", "GSM2371014", "GSM2371015", "GSM2371016", "GSM2371017", "GSM2371018", "GSM2371019", "GSM2371020", "GSM2371021", "GSM2371022", "GSM2371023", "GSM2371024", "GSM2371025", "GSM2371026", "GSM2371027", "GSM2371028", "GSM2371029", "GSM2371030", "GSM2371031", "GSM2371032", "GSM2371033", "GSM2371034", "GSM2371035", "GSM2371036", "GSM2371037", "GSM2371038", "GSM2371039", "GSM2371040", "GSM2371041", "GSM2371042", "GSM2371043", "GSM2371044", "GSM2371045", "GSM2371046", "GSM2371047", "GSM2371048", "GSM2371049", "GSM2371050", "GSM2371051", "GSM2371052", "GSM2371053", "GSM2371054", "GSM2371055", "GSM2371056", "GSM2371057", "GSM2371058", "GSM2371059", "GSM2371060", "GSM2371061", "GSM2371062", "GSM2371063", "GSM2371064", "GSM2371065", "GSM2371066", "GSM2371067", "GSM2371068", "GSM2371069", "GSM2371070", "GSM2371071", "GSM2371072", "GSM2371073", "GSM2371074", "GSM2371075", "GSM2371076", "GSM2371077", "GSM2371078", "GSM2371079", "GSM2371080", "GSM2371081", "GSM2371082", "GSM2371083", "GSM2371084", "GSM2371085", "GSM2371086", "GSM2371087", "GSM2371088", "GSM2371089", "GSM2371090", "GSM2371091", "GSM2371092", "GSM2371093", "GSM2371094", "GSM2371095", "GSM2371096", "GSM2371097", "GSM2371098", "GSM2371099", "GSM2371100", "GSM2371101", "GSM2371102", "GSM2371103", "GSM2371104", "GSM2371105", "GSM2371106", "GSM2371107", "GSM2371108", "GSM2371109", "GSM2371110", "GSM2371111", "GSM2371112", "GSM2371113", "GSM2371114", "GSM2371115", "GSM2371116", "GSM2371117", "GSM2371118", "GSM2371119", "GSM2371120", "GSM2371121", "GSM2371122", "GSM2371123", "GSM2371124", "GSM2371125", "GSM2371126", "GSM2371127", "GSM2371128", "GSM2371129", "GSM2371130", "GSM2371131", "GSM2371132", "GSM2371133", "GSM2371134", "GSM2371135", "GSM2371136", "GSM2371137", "GSM2371138", "GSM2371139", "GSM2371140", "GSM2371141", "GSM2371142", "GSM2371143", "GSM2371144", "GSM2371145", "GSM2371146", "GSM2371147", "GSM2371148", "GSM2371149", "GSM2371150", "GSM2371151", "GSM2371152", "GSM2371153", "GSM2371154", "GSM2371155", "GSM2371156", "GSM2371157", "GSM2371158", "GSM2371159", "GSM2371160", "GSM2371161", "GSM2371162", "GSM2371163", "GSM2371164", "GSM2371165", "GSM2371166", "GSM2371167", "GSM2371168", "GSM2371169", "GSM2371170", "GSM2371171", "GSM2371172", "GSM2371173", "GSM2371174", "GSM2371175", "GSM2371176", "GSM2371177", "GSM2371178", "GSM2371179", "GSM2371180", "GSM2371181", "GSM2371182", "GSM2371183", "GSM2371184", "GSM2371185", "GSM2371186", "GSM2371187"], "sample_count": 218, "group_sizes": null, "group_sizes_basis": null, "platform": "11154", "taxon": "Homo sapiens", "title": "RNAseq study of synovial biopsies", "resolution_checked_at": "2026-11-01T15:39:55Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}, {"id": "data:geo:GSE90081", "role": "raw", "format": "geo-series", "media_type": "application/vnd.ncbi.geo-series+json", "uri": "https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE90081", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE90081", "version": null, "note": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21].", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21].", "access_statement_kind": "repository_route_statement", "sample_accessions": ["GSM2397368", "GSM2397369", "GSM2397370", "GSM2397371", "GSM2397372", "GSM2397373", "GSM2397374", "GSM2397375", "GSM2397376", "GSM2397377", "GSM2397378", "GSM2397379", "GSM2397380", "GSM2397381", "GSM2397382", "GSM2397383", "GSM2397384", "GSM2397385", "GSM2397386", "GSM2397387", "GSM2397388", "GSM2397389", "GSM2397390", "GSM2397391"], "sample_count": 24, "group_sizes": null, "group_sizes_basis": null, "platform": "11154", "taxon": "Homo sapiens", "title": "Discovery of new candidate genes for rheumatoid arthritis through integration of genetic association data with expression pathway analysis", "resolution_checked_at": "2026-11-01T20:16:08Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}], "gaps": ["asset:data:geo:GSE89408 not downloaded", "asset:data:geo:GSE89408 licence not read", "asset:data:geo:GSE90081 not downloaded", "asset:data:geo:GSE90081 licence not read"]}, "analysis": {"state": "partial", "methods": [{"id": "method-1", "family": "machine_learning", "specific": "random_forest", "label": "Machine learning", "dsl_operations": [], "missing_operations": ["machine_learning_random_forest"], "expressible": false, "parameters": {}, "data_shapes": ["labelled_feature_matrix"], "evidence": {"score": 1.0, "band": "verbatim", "quote": "We applied machine learning algorithms, including LASSO, random forest (RF), and support vector machine-recursive feature elimination (SVM-RFE), to identify the key genes, discern the hub genes, assess their diagnostic potential, and correlate them with immune cell infiltration patterns."}}, {"id": "method-2", "family": "machine_learning", "specific": "support_vector_machine", "label": "Machine learning", "dsl_operations": ["train_classifier", "cross_validate", "roc_auc"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["labelled_feature_matrix"], "evidence": {"score": 0.7, "band": "medium", "quote": "support vector machine"}}, {"id": "method-3", "family": "network_analysis", "specific": "ppi_network", "label": "Network / pathway analysis", "dsl_operations": ["build_graph", "centrality_scores", "pathway_enrich"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["edge_list"], "evidence": {"score": 0.7, "band": "medium", "quote": "protein-protein interaction"}}, {"id": "method-4", "family": "diagnostic_evaluation", "specific": "roc", "label": "Diagnostic test evaluation", "dsl_operations": ["confusion_counts", "roc_curve", "sensitivity_specificity", "solve_confusion_matrix", "classification_metrics", "published_precision_envelope"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["reference_standard_and_test"], "evidence": {"score": 0.7, "band": "medium", "quote": "ROC curve"}}], "targets": [], "gaps": ["method.machine_learning_random_forest"]}, "capability": {"state": "complete", "requirements": [{"capability_id": "capability:computational_analysis", "label": "Computational analysis platform", "assay_families": ["meta_analysis", "machine_learning", "survival_analysis", "statistical_regression", "pca", "molecular_docking", "network_analysis", "differential_expression", "gwas"], "availability_state": "owned", "tier_unlocked": "R1", "candidate": "Automated Assays Laravel application and simulated adapters"}], "gaps": []}, "readiness": {"state": "complete", "tier": "R1", "blocking_axis": "none", "blocked_on": "nothing:replicable_now", "axes": {"method": {"tier": "R1", "expressible": false, "required_operations": [], "missing_operations": ["machine_learning_random_forest"], "basis": "The paper names the method it used and the corpus's method extractor maps it onto the DSL catalogue: random_forest."}, "capability": {"tier": "R1", "capability_id": "capability:computational_analysis", "availability_state": "owned", "basis": "The computational analysis is owned. The physical bench the paper used is not, so a physical validation cannot be run here."}, "data": {"tier": "R1", "reachable": true, "accessions": ["GSE89408", "GSE90081"], "resolved_accessions": ["GSE89408", "GSE90081"], "statements": ["The bulk RNA dataset GSE89408 was retrieved from the GEO database (https://www.ncbi.nlm.nih.gov/geo/), based on the GPL11154 Illumina HiSeq 2000 platform. This dataset comprised 218 samples, including 152 RA (46 males and 106 females) samples and 28 healthy control tissue samples (14 males and 14 females) from the synovium, which were utilized further for this study [21]."], "basis": "The paper names GSE89408, GSE90081 in Gene Expression Omnibus (GEO) and the accession resolves in the repository. No dataset was downloaded and no licence was read."}}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:41021591", "data:geo:GSE89408"], "steps": []}, "must_build": {"dsl_operations": [{"operation": "machine_learning_random_forest", "unlocks_papers": null, "priority": null}], "converters": [], "readers": [], "capabilities": [], "data_gaps": ["data:geo:GSE89408"]}, "gaps": []}, "review": {"state": "partial", "machine_critique": {"kind": "extraction_review", "reader": {"model": "MiniMax-M3.1-Flash", "provider": "minimax", "provider_family": "minimax", "read_at": "2026-11-01T20:02:12Z", "independence": "single_reader"}, "evidence_check": {"method": "verbatim", "availability": null}, "calls": 5, "usage": {"prompt_tokens": 12887, "completion_tokens": 2085, "cached_tokens": 128}}, "human_critique": null, "stages": [{"id": "extraction", "state": "complete", "reader": "MiniMax-M3.1-Flash", "provider": "minimax"}], "gaps": ["review.independent_reader", "review.machine_critique", "review.human_critique"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "pmid:41021591", "type": "paper", "label": "pmid:41021591"}, {"id": "source:pmid:41021591", "type": "manuscript", "label": "PMID 41021591 full text"}, {"id": "data:geo:GSE89408", "type": "dataset", "label": "GEO GSE89408"}], "activities": [{"id": "activity-reading-sci-2026-11-01", "type": "classification", "used": [{"id": "pmid:41021591", "type": "paper", "label": "pmid:41021591"}, {"id": "source:pmid:41021591", "type": "manuscript", "label": "PMID 41021591 full text"}, {"id": "data:geo:GSE89408", "type": "dataset", "label": "GEO GSE89408"}], "agent": "agent:corpus-assay-engine"}, {"id": "activity-read-41021591", "type": "extraction", "used": ["source:pmid:41021591"], "agent": "agent:minimax-reader"}, {"id": "activity-resolve-GSE89408", "type": "accession_resolution", "used": ["data:geo:GSE89408"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}, {"id": "agent:minimax-reader", "type": "software", "label": "MiniMax literature reader"}], "source_run_id": "reading-sci-2026-11-01", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "pmid:41021591", "parent_hash": "sha256:aaa3b16cbcca812f437b8c2a07df6eaf6b8b7718ae25000f0bd9c78c20bbfa32", "fork_of": null, "generation": 3}, "gaps": ["asset:data:geo:GSE89408 licence not read", "asset:data:geo:GSE89408 not downloaded", "asset:data:geo:GSE90081 licence not read", "asset:data:geo:GSE90081 not downloaded", "execution.not_run", "method.machine_learning_random_forest", "review.human_critique", "review.independent_reader", "review.machine_critique", "step:run_analysis blocked on method:machine_learning_random_forest"], "assays": [{"role": "analysis", "definition_id": "machine_learning", "definition_version": "f2132e96ae6c7a4e", "definition_digest": null, "required_operations": [], "support_state": "partial", "source_hashes": {"dsl_catalogue": "f2132e96ae6c7a4e"}, "confidence": 1.0, "evidence": ["We applied machine learning algorithms, including LASSO, random forest (RF), and support vector machine-recursive feature elimination (SVM-RFE), to identify the key genes, discern the hub genes, assess their diagnostic potential, and correlate them with immune cell infiltration patterns."], "order_basis": "analysis_only", "id": "assay:1"}]}