Deciphering the role of lipid metabolism-related genes in Alzheimer's disease: a machine learning approach integrating Traditional Chinese Medicine.

{"sci_version": "1.2", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:39507054", "kind": "science-record", "revision": 4, "content_hash": "sha256:e6e96b7f65c2e4be05c35b7074d5a3d79f5d773bd0634c9e57816c3fffdab1b6", "created_at": "2026-11-01T20:29:45Z", "updated_at": "2026-11-01T22:55:33Z", "generator": {"name": "corpus-assay-engine", "run_id": "reading-sci-2026-11-01"}, "paper": {"state": "complete", "title": "Deciphering the role of lipid metabolism-related genes in Alzheimer's disease: a machine learning approach integrating Traditional Chinese Medicine.", "doi": "10.3389/fendo.2024.1448119", "pmid": "39507054", "authors": ["KeShangJing Wu", "QingSong Liu", "KeYu Long", "XueQing Duan", "XianYu Chen", "Jing Zhang", "Li Li", "Bin Li"], "journal": "Frontiers in endocrinology", "year": 2024, "abstract": "Background: Alzheimer's disease (AD) represents a progressive neurodegenerative disorder characterized by the accumulation of misfolded amyloid beta protein, leading to the formation of amyloid plaques and the aggregation of tau protein into neurofibrillary tangles within the cerebral cortex. The role of carbohydrates, particularly apolipoprotein E (ApoE), is pivotal in AD pathogenesis due to its involvement in lipid and cholesterol metabolism, and its status as a genetic predisposition factor for the disease. Despite its significance, the mechanistic contributions of Lipid Metabolism-related Genes (LMGs) to AD remain inadequately elucidated. This research endeavor seeks to bridge this gap by pinpointing biomarkers indicative of early-stage AD, with an emphasis on those linked to immune cell infiltration. To this end, advanced machine-learning algorithms and data derived from the Gene Expression Omnibus (GEO) database have been employed to facilitate the identification of these biomarkers.\nMethods: Differentially expressed genes (DEGs) were identified by comparing gene expression profiles between healthy individuals and Alzheimer's disease (AD) patients, using data from two Gene Expression Omnibus (GEO) datasets: GSE5281 and GSE138260. Functional enrichment analysis was conducted to elucidate the biological relevance of the DEGs. To ensure the reliability of the results, samples were randomly divided into training and validation sets. The analysis focused on lipid metabolism-related DEGs (LMDEGs) to explore potential biomarkers for AD. Machine learning algorithms, including Support Vector Machine-Recursive Feature Elimination (SVM-RFE) and the Least Absolute Shrinkage and Selection Operator (LASSO) regression model, were applied to identify a key gene biomarker. Additionally, immune cell infiltration and its relationship with the gene biomarker were assessed using the CIBERSORT algorithm. The Integrated Traditional Chinese Medicine (ITCM) database was also referenced to identify Chinese medicines related to lipid metabolism and their possible connection to AD. This comprehensive strategy aims to integrate modern computational methods with traditional medicine to deepen our understanding of AD and its underlying mechanisms.\nResults: The study identified 137 genes from a pool of 751 lipid metabolism-related genes (LMGs) significantly associated with autophagy and immune response mechanisms. Through the application of LASSO and SVM-RFE machine-learning techniques, four genes-choline acetyl transferase (CHAT), member RAS oncogene family (RAB4A), acyl-CoA binding domain-containing protein 6 (ACBD6), and alpha-galactosidase A (GLA)-emerged as potential biomarkers for Alzheimer's disease (AD). These genes demonstrated strong therapeutic potential due to their involvement in critical biological pathways. Notably, nine Chinese medicine compounds were identified to target these marker genes, offering a novel treatment approach for AD. Further, ceRNA network analysis revealed complex regulatory interactions involving these genes, underscoring their importance in AD pathology. CIBERSORT analysis highlighted a potential link between changes in the immune microenvironment and CHAT expression levels in AD patients, providing new insights into the immunological dimensions of the disease.\nConclusion: The discovery of these gene markers offers substantial promise for the diagnosis and understanding of Alzheimer's disease (AD). However, further investigation is necessary to validate their clinical utility. This study illuminates the role of Lipid Metabolism-related Genes (LMGs) in AD pathogenesis, offering potential targets for therapeutic intervention. It enhances our grasp of AD's complex mechanisms and paves the way for future research aimed at refining diagnostic and treatment strategies.", "claims": [{"id": "claim-method", "statement": "SVM-RFE filtered the DEGs to identify the optimal combination of feature genes, yielding four marker genes (CHAT, RAB4A, ACBD6, and alpha-galactosidase A [GLA]) for subsequent analysis ( Figure 3E ).", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:39507054", "locator": "sections[16] \"Four DEGs identified as AD diagnostic genes\" paragraph 2", "quote": "SVM-RFE filtered the DEGs to identify the optimal combination of feature genes, yielding four marker genes (CHAT, RAB4A, ACBD6, and alpha-galactosidase A [GLA]) for subsequent analysis ( Figure 3E )."}], "state": "complete"}, {"id": "claim-data-access", "statement": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:39507054", "locator": "sections[27] \"Data availability statement\" paragraph 2", "quote": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors."}], "state": "complete"}], "gaps": []}, "experiment": {"state": "complete", "objective": null, "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:39507054"], "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:GSE138260"], "outputs": ["the source data with its repository provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the published analysis", "kind": "analysis", "state": "complete", "blocked_on": null, "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:GSE138260"], "outputs": ["the replicated result and its comparison"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": []}, "data": {"state": "partial", "assets": [{"id": "source:pmid:39507054", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.3389/fendo.2024.1448119", "sha256": "sha256:48b166f5787526d689ade226c11a55d912aa4c7f215de478ff3a0325c8a723db", "bytes": 58682, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.3389/fendo.2024.1448119", "version": null, "note": "Full article fetched from PMC; the checksum covers the rendered section body, not the original XML bytes.", "structure": {"pmcid": "PMC11538058", "source": "europe_pmc_fulltext", "coverage": "whole_body", "text_hash_scope": "rendered_section_body"}}, {"id": "data:geo:GSE138260", "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=GSE138260", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE138260", "version": null, "note": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.", "access_statement_kind": "data_availability_statement", "sample_accessions": ["GSM4103849", "GSM4103850", "GSM4103851", "GSM4103852", "GSM4103853", "GSM4103854", "GSM4103855", "GSM4103856", "GSM4103857", "GSM4103858", "GSM4103859", "GSM4103860", "GSM4103861", "GSM4103862", "GSM4103863", "GSM4103864", "GSM4103865", "GSM4103866", "GSM4103867", "GSM4103868", "GSM4103869", "GSM4103870", "GSM4103871", "GSM4103872", "GSM4103873", "GSM4103874", "GSM4103875", "GSM4103876", "GSM4103877", "GSM4103878", "GSM4103879", "GSM4103880", "GSM4103881", "GSM4103882", "GSM4103883", "GSM4103884"], "sample_count": 36, "group_sizes": null, "group_sizes_basis": null, "platform": "27556", "taxon": "Homo sapiens", "title": "Differential expression analysis of human brain samples with Alzheimer disease versus healthy control samples", "resolution_checked_at": "2026-11-01T18:46:34Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}, {"id": "data:geo:GSE5281", "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=GSE5281", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE5281", "version": null, "note": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors.", "access_statement_kind": "data_availability_statement", "sample_accessions": ["GSM119615", "GSM119616", "GSM119617", "GSM119618", "GSM119619", "GSM119620", "GSM119621", "GSM119622", "GSM119623", "GSM119624", "GSM119625", "GSM119626", "GSM119627", "GSM119628", "GSM119629", "GSM119630", "GSM119631", "GSM119632", "GSM119633", "GSM119634", "GSM119635", "GSM119636", "GSM119637", "GSM119638", "GSM119639", "GSM119640", "GSM119641", "GSM119642", "GSM119643", "GSM119644", "GSM119645", "GSM119646", "GSM119647", "GSM119648", "GSM119649", "GSM119650", "GSM119651", "GSM119652", "GSM119653", "GSM119654", "GSM119655", "GSM119656", "GSM119657", "GSM119658", "GSM119659", "GSM119660", "GSM119661", "GSM119662", "GSM119663", "GSM119664", "GSM119665", "GSM119666", "GSM119667", "GSM119668", "GSM119669", "GSM119670", "GSM119671", "GSM119672", "GSM119673", "GSM119674", "GSM119675", "GSM119676", "GSM119677", "GSM119678", "GSM119679", "GSM119680", "GSM119681", "GSM119682", "GSM119683", "GSM119684", "GSM119685", "GSM119686", "GSM119687", "GSM119688", "GSM238763", "GSM238790", "GSM238791", "GSM238792", "GSM238793", "GSM238794", "GSM238795", "GSM238796", "GSM238797", "GSM238798", "GSM238799", "GSM238800", "GSM238801", "GSM238802", "GSM238803", "GSM238804", "GSM238805", "GSM238806", "GSM238807", "GSM238808", "GSM238809", "GSM238810", "GSM238811", "GSM238812", "GSM238813", "GSM238815", "GSM238816", "GSM238817", "GSM238818", "GSM238819", "GSM238820", "GSM238821", "GSM238822", "GSM238823", "GSM238824", "GSM238825", "GSM238826", "GSM238827", "GSM238834", "GSM238835", "GSM238837", "GSM238838", "GSM238839", "GSM238840", "GSM238841", "GSM238842", "GSM238843", "GSM238844", "GSM238845", "GSM238846", "GSM238847", "GSM238848", "GSM238851", "GSM238854", "GSM238855", "GSM238856", "GSM238857", "GSM238858", "GSM238860", "GSM238861", "GSM238862", "GSM238863", "GSM238864", "GSM238865", "GSM238867", "GSM238868", "GSM238870", "GSM238871", "GSM238872", "GSM238873", "GSM238874", "GSM238875", "GSM238877", "GSM238941", "GSM238942", "GSM238943", "GSM238944", "GSM238945", "GSM238946", "GSM238947", "GSM238948", "GSM238949", "GSM238951", "GSM238952", "GSM238953", "GSM238955", "GSM238963"], "sample_count": 161, "group_sizes": null, "group_sizes_basis": null, "platform": "570", "taxon": "Homo sapiens", "title": "Alzheimer's disease and the normal aged brain (steph-affy-human-433773)", "resolution_checked_at": "2026-11-01T15:39:54Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}], "gaps": ["asset:data:geo:GSE138260 not downloaded", "asset:data:geo:GSE138260 licence not read", "asset:data:geo:GSE5281 not downloaded", "asset:data:geo:GSE5281 licence not read"]}, "analysis": {"state": "complete", "methods": [{"id": "method-1", "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": 1.0, "band": "verbatim", "quote": "SVM-RFE filtered the DEGs to identify the optimal combination of feature genes, yielding four marker genes (CHAT, RAB4A, ACBD6, and alpha-galactosidase A [GLA]) for subsequent analysis ( Figure 3E )."}}, {"id": "method-2", "family": "correlation_analysis", "specific": "pearson", "label": "Correlation analysis", "dsl_operations": ["pearson_r"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["paired_numeric_columns"], "evidence": {"score": 0.8, "band": "high", "quote": "Pearson correlation"}}], "targets": [], "gaps": []}, "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": true, "required_operations": ["train_classifier", "cross_validate", "roc_auc"], "missing_operations": [], "basis": "The paper names the method it used and the corpus's method extractor maps it onto the DSL catalogue: support_vector_machine."}, "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": ["GSE138260", "GSE5281"], "resolved_accessions": ["GSE138260", "GSE5281"], "statements": ["The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authors."], "basis": "The paper names GSE138260, GSE5281 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": ["train_classifier", "cross_validate", "roc_auc"], "assets": ["source:pmid:39507054", "data:geo:GSE138260"], "steps": ["run_analysis"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": [], "capabilities": [], "data_gaps": ["data:geo:GSE138260"]}, "gaps": []}, "review": {"state": "partial", "machine_critique": {"kind": "extraction_review", "reader": {"model": "MiniMax-M3.1-Flash", "provider": "minimax", "provider_family": "minimax", "read_at": "2026-11-01T19:37:44Z", "independence": "single_reader"}, "evidence_check": {"method": "verbatim", "availability": "verbatim"}, "calls": 5, "usage": {"prompt_tokens": 15761, "completion_tokens": 1261, "cached_tokens": 188}}, "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:39507054", "type": "paper", "label": "pmid:39507054"}, {"id": "source:pmid:39507054", "type": "manuscript", "label": "PMID 39507054 full text"}, {"id": "data:geo:GSE138260", "type": "dataset", "label": "GEO GSE138260"}], "activities": [{"id": "activity-reading-sci-2026-11-01", "type": "classification", "used": [{"id": "pmid:39507054", "type": "paper", "label": "pmid:39507054"}, {"id": "source:pmid:39507054", "type": "manuscript", "label": "PMID 39507054 full text"}, {"id": "data:geo:GSE138260", "type": "dataset", "label": "GEO GSE138260"}], "agent": "agent:corpus-assay-engine"}, {"id": "activity-read-39507054", "type": "extraction", "used": ["source:pmid:39507054"], "agent": "agent:minimax-reader"}, {"id": "activity-resolve-GSE138260", "type": "accession_resolution", "used": ["data:geo:GSE138260"], "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:39507054", "parent_hash": "sha256:be899c6e2e2636afcdf6b50d4a151b03f1a26770d1c71fd10e2741324fd5ea0f", "fork_of": null, "generation": 3}, "gaps": ["asset:data:geo:GSE138260 licence not read", "asset:data:geo:GSE138260 not downloaded", "asset:data:geo:GSE5281 licence not read", "asset:data:geo:GSE5281 not downloaded", "execution.not_run", "review.human_critique", "review.independent_reader", "review.machine_critique"], "assays": [{"role": "analysis", "definition_id": "machine_learning", "definition_version": "f2132e96ae6c7a4e", "definition_digest": null, "required_operations": ["train_classifier", "cross_validate", "roc_auc"], "support_state": "supported", "source_hashes": {"dsl_catalogue": "f2132e96ae6c7a4e"}, "confidence": 1.0, "evidence": ["SVM-RFE filtered the DEGs to identify the optimal combination of feature genes, yielding four marker genes (CHAT, RAB4A, ACBD6, and alpha-galactosidase A [GLA]) for subsequent analysis ( Figure 3E )."], "order_basis": "analysis_only", "id": "assay:1"}]}