Integrative analysis of DNA methylation and gene expression through machine learning identifies stomach cancer diagnostic and prognostic biomarkers.

{"sci_version": "1.2", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:36779430", "kind": "science-record", "revision": 3, "content_hash": "sha256:eb54b4f8c4eebc0e909f3618ba9c76048db98b3e06ffca6294127a517fc7261f", "created_at": "2026-11-01T19:01:22Z", "updated_at": "2026-11-01T22:08:48Z", "generator": {"name": "corpus-assay-engine", "run_id": "reading-sci-2026-11-01"}, "paper": {"state": "complete", "title": "Integrative analysis of DNA methylation and gene expression through machine learning identifies stomach cancer diagnostic and prognostic biomarkers.", "doi": "10.1111/jcmm.17693", "pmid": "36779430", "authors": ["Maryam Hosseini", "Maryam Lotfi-Shahreza", "Parvaneh Nikpour"], "journal": "Journal of cellular and molecular medicine", "year": 2023, "abstract": "DNA methylation is an early event in tumorigenesis. Here, by integrative analysis of DNA methylation and gene expression and utilizing machine learning approaches, we introduced potential diagnostic and prognostic methylation signatures for stomach cancer. Differentially-methylated positions (DMPs) and differentially-expressed genes (DEGs) were identified using The Cancer Genome Atlas (TCGA) stomach adenocarcinoma (STAD) data. A total of 256 DMPs consisting of 140 and 116 hyper- and hypomethylated positions were identified between 443 tumour and 27 nontumour STAD samples. Gene expression analysis revealed a total of 2821 DEGs with 1247 upregulated and 1574 downregulated genes. By analysing the impact of cis and trans regulation of methylation on gene expression, a dominant negative correlation between methylation and expression was observed, while for trans regulation, in hypermethylated and hypomethylated genes, there was mainly a negative and positive correlation with gene expression, respectively. To find diagnostic biomarkers, we used 28 hypermethylated probes locating in the promoter of 27 downregulated genes. By implementing a feature selection approach, eight probes were selected and then used to build a support vector machine diagnostic model, which had an area under the curve of 0.99 and 0.97 in the training and validation (GSE30601 with 203 tumour and 94 nontumour samples) cohorts, respectively. Using 412 TCGA-STAD samples with both methylation and clinical data, we also identified four prognostic probes by implementing univariate and multivariate Cox regression analysis. In summary, our study introduced potential diagnostic and prognostic biomarkers for STAD, which demands further validation.", "claims": [{"id": "claim-method", "statement": "Recursive feature elimination with cross‐validation (RFECV) method 20 was then implemented on the remaining probes to select final features for model construction.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:36779430", "locator": "sections[7] \"Identification of candidate diagnostic CpG probes, model construction and evaluation\" paragraph 2", "quote": "Recursive feature elimination with cross‐validation (RFECV) method 20 was then implemented on the remaining probes to select final features for model construction."}], "state": "complete"}, {"id": "claim-data-access", "statement": "The codes used to perform the analysis are available in the Mendeley Data repository, at https://doi.org/10.17632/dcdy79pjcw.1. The results published or shown here are in whole or part based on data generated by the TCGA Research Network (https://portal.gdc.cancer.gov/legacy‐archive/search/f) and GEO (https://www.ncbi.nlm.nih.gov/geo/). GEO data can be accessed with the unique identifier GSE30601.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:36779430", "locator": "sections[26] \"DATA AVAILABILITY STATEMENT\" paragraph 2", "quote": "The codes used to perform the analysis are available in the Mendeley Data repository, at https://doi.org/10.17632/dcdy79pjcw.1. The results published or shown here are in whole or part based on data generated by the TCGA Research Network (https://portal.gdc.cancer.gov/legacy‐archive/search/f) and GEO (https://www.ncbi.nlm.nih.gov/geo/). GEO data can be accessed with the unique identifier GSE30601."}], "state": "complete"}, {"id": "claim-target-result", "statement": "The study identified eight diagnostic methylation probes and four prognostic methylation probes for stomach adenocarcinoma, with the diagnostic SVM model achieving an AUC of 0.97 in an independent validation set.", "kind": "qualitative", "target": null, "evidence": [{"asset": "source:pmid:36779430", "locator": "sections[1] \"INTRODUCTION\" paragraph 4", "quote": "In the current study, by integrated analysis of methylation, RNA‐seq gene expression and clinical data of stomach adenocarcinoma (STAD) from The Cancer Genome Atlas (TCGA) database and applying a machine learning approach, we identified eight potential diagnostic methylation biomarkers with a precision of 0.94, recall of 0.93 and area under the curve (AUC) of 0.97 in an independent validation set from the Gene Expression Omnibus (GEO) database."}], "state": "complete"}], "gaps": []}, "experiment": {"state": "complete", "objective": "The study identified eight diagnostic methylation probes and four prognostic methylation probes for stomach adenocarcinoma, with the diagnostic SVM model achieving an AUC of 0.97 in an independent validation set.", "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:36779430"], "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:GSE30601"], "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:GSE30601"], "outputs": ["the replicated result and its comparison"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": []}, "data": {"state": "partial", "assets": [{"id": "source:pmid:36779430", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.1111/jcmm.17693", "sha256": "sha256:662526c04321d190dfd692bb17239a166f3492c89760e810fcabc4c508a31ccd", "bytes": 49343, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.1111/jcmm.17693", "version": null, "note": "Full article fetched from PMC; the checksum covers the rendered section body, not the original XML bytes.", "structure": {"pmcid": "PMC9983314", "source": "europe_pmc_fulltext", "coverage": "whole_body", "text_hash_scope": "rendered_section_body"}}, {"id": "data:geo:GSE30601", "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=GSE30601", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE30601", "version": null, "note": "The codes used to perform the analysis are available in the Mendeley Data repository, at https://doi.org/10.17632/dcdy79pjcw.1. The results published or shown here are in whole or part based on data generated by the TCGA Research Network (https://portal.gdc.cancer.gov/legacy‐archive/search/f) and GEO (https://www.ncbi.nlm.nih.gov/geo/). GEO data can be accessed with the unique identifier GSE30601.", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The codes used to perform the analysis are available in the Mendeley Data repository, at https://doi.org/10.17632/dcdy79pjcw.1. The results published or shown here are in whole or part based on data generated by the TCGA Research Network (https://portal.gdc.cancer.gov/legacy‐archive/search/f) and GEO (https://www.ncbi.nlm.nih.gov/geo/). GEO data can be accessed with the unique identifier GSE30601.", "access_statement_kind": "data_availability_statement", "sample_accessions": ["GSM759072", "GSM759073", "GSM759074", "GSM759075", "GSM759076", "GSM759077", "GSM759078", "GSM759079", "GSM759080", "GSM759081", "GSM759082", "GSM759083", "GSM759084", "GSM759085", "GSM759086", "GSM759087", "GSM759088", "GSM759089", "GSM759090", "GSM759091", "GSM759092", "GSM759093", "GSM759094", "GSM759095", "GSM759096", "GSM759097", "GSM759098", "GSM759099", "GSM759100", "GSM759101", "GSM759102", "GSM759103", "GSM759104", "GSM759105", "GSM759106", "GSM759107", "GSM759108", "GSM759109", "GSM759110", "GSM759111", "GSM759112", "GSM759113", "GSM759114", "GSM759115", "GSM759116", "GSM759117", "GSM759118", "GSM759119", "GSM759120", "GSM759121", "GSM759122", "GSM759123", "GSM759124", "GSM759125", "GSM759126", "GSM759127", "GSM759128", "GSM759129", "GSM759130", "GSM759131", "GSM759132", "GSM759133", "GSM759134", "GSM759135", "GSM759136", "GSM759137", "GSM759138", "GSM759139", "GSM759140", "GSM759141", "GSM759142", "GSM759143", "GSM759144", "GSM759145", "GSM759146", "GSM759147", "GSM759148", "GSM759149", "GSM759150", "GSM759151", "GSM759152", "GSM759153", "GSM759154", "GSM759155", "GSM759156", "GSM759157", "GSM759158", "GSM759159", "GSM759160", "GSM759161", "GSM759162", "GSM759163", "GSM759164", "GSM759165", "GSM759166", "GSM759167", "GSM759168", "GSM759169", "GSM759170", "GSM759171", "GSM759172", "GSM759173", "GSM759174", "GSM759175", "GSM759176", "GSM759177", "GSM759178", "GSM759179", "GSM759180", "GSM759181", "GSM759182", "GSM759183", "GSM759184", "GSM759185", "GSM759186", "GSM759187", "GSM759188", "GSM759189", "GSM759190", "GSM759191", "GSM759192", "GSM759193", "GSM759194", "GSM759195", "GSM759196", "GSM759197", "GSM759198", "GSM759199", "GSM759200", "GSM759201", "GSM759202", "GSM759203", "GSM759204", "GSM759205", "GSM759206", "GSM759207", "GSM759208", "GSM759209", "GSM759210", "GSM759211", "GSM759212", "GSM759213", "GSM759214", "GSM759215", "GSM759216", "GSM759217", "GSM759218", "GSM759219", "GSM759220", "GSM759221", "GSM759222", "GSM759223", "GSM759224", "GSM759225", "GSM759226", "GSM759227", "GSM759228", "GSM759229", "GSM759230", "GSM759231", "GSM759232", "GSM759233", "GSM759234", "GSM759235", "GSM759236", "GSM759237", "GSM759238", "GSM759239", "GSM759240", "GSM759241", "GSM759242", "GSM759243", "GSM759244", "GSM759245", "GSM759246", "GSM759247", "GSM759248", "GSM759249", "GSM759250", "GSM759251", "GSM759252", "GSM759253", "GSM759254", "GSM759255", "GSM759256", "GSM759257", "GSM759258", "GSM759259", "GSM759260", "GSM759261", "GSM759262", "GSM759263", "GSM759264", "GSM759265", "GSM759266", "GSM759267", "GSM759268", "GSM759269", "GSM759270", "GSM759271", "GSM759272", "GSM759273", "GSM759274", "GSM759275", "GSM759276", "GSM759277", "GSM759278", "GSM759279", "GSM759280", "GSM759281", "GSM759282", "GSM759283", "GSM759284", "GSM759285", "GSM759286", "GSM759287", "GSM759288", "GSM759289", "GSM759290", "GSM759291", "GSM759292", "GSM759293", "GSM759294", "GSM759295", "GSM759296", "GSM759297", "GSM759298", "GSM759299", "GSM759300", "GSM759301", "GSM759302", "GSM759303", "GSM759304", "GSM759305", "GSM759306", "GSM759307", "GSM759308", "GSM759309", "GSM759310", "GSM759311", "GSM759312", "GSM759313", "GSM759314", "GSM759315", "GSM759316", "GSM759317", "GSM759318", "GSM759319", "GSM759320", "GSM759321", "GSM759322", "GSM759323", "GSM759324", "GSM759325", "GSM759326", "GSM759327", "GSM759328", "GSM759329", "GSM759330", "GSM759331", "GSM759332", "GSM759333", "GSM759334", "GSM759335", "GSM759336", "GSM759337", "GSM759338", "GSM759339", "GSM759340", "GSM759341", "GSM759342", "GSM759343", "GSM759344", "GSM759345", "GSM759346", "GSM759347", "GSM759348", "GSM759349", "GSM759350", "GSM759351", "GSM759352", "GSM759353", "GSM759354", "GSM759355", "GSM759356", "GSM759357", "GSM759358", "GSM759359", "GSM759360", "GSM759361", "GSM759362", "GSM759363", "GSM759364", "GSM759365", "GSM759366", "GSM759367", "GSM759368"], "sample_count": 297, "group_sizes": null, "group_sizes_basis": null, "platform": "8490", "taxon": "Homo sapiens", "title": "Epigenetic analysis of gastric cancer", "resolution_checked_at": "2026-11-01T17:30:42Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}], "gaps": ["asset:data:geo:GSE30601 not downloaded", "asset:data:geo:GSE30601 licence not read"]}, "analysis": {"state": "partial", "methods": [{"id": "method-1", "family": "survival_analysis", "specific": "cox_proportional_hazards", "label": "Survival analysis", "dsl_operations": ["cox_ph"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["time_to_event_with_censoring"], "evidence": {"score": 0.7, "band": "medium", "quote": "Cox regression"}}, {"id": "method-2", "family": "differential_expression", "specific": "deseq2", "label": "Differential expression analysis", "dsl_operations": ["normalise_counts", "differential_expression_fit", "fdr_bh"], "missing_operations": [], "expressible": true, "parameters": {}, "data_shapes": ["count_matrix_with_condition"], "evidence": {"score": 0.8, "band": "high", "quote": "DESeq2"}}, {"id": "method-3", "family": "machine_learning", "specific": "support_vector_machine", "label": "Machine learning", "dsl_operations": ["train_classifier", "cross_validate", "roc_auc"], "missing_operations": [], "expressible": true, "parameters": {"auc": "tic model, which had an area under the curve of 0"}, "data_shapes": ["labelled_feature_matrix"], "evidence": {"score": 0.7, "band": "medium", "quote": "support vector machine"}}, {"id": "method-4", "family": null, "specific": null, "label": "Recursive feature elimination with cross‐validation (RFECV)", "dsl_operations": [], "missing_operations": [], "expressible": null, "parameters": {}, "data_shapes": [], "evidence": {"score": 1.0, "band": "verbatim", "quote": "Recursive feature elimination with cross‐validation (RFECV) method 20 was then implemented on the remaining probes to select final features for model construction."}}], "targets": [], "gaps": ["method.reader_named_unmapped"]}, "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": ["cox_ph"], "missing_operations": [], "basis": "The paper names the method it used and the corpus's method extractor maps it onto the DSL catalogue: cox_proportional_hazards."}, "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": ["GSE30601"], "resolved_accessions": ["GSE30601"], "statements": ["The codes used to perform the analysis are available in the Mendeley Data repository, at https://doi.org/10.17632/dcdy79pjcw.1. The results published or shown here are in whole or part based on data generated by the TCGA Research Network (https://portal.gdc.cancer.gov/legacy‐archive/search/f) and GEO (https://www.ncbi.nlm.nih.gov/geo/). GEO data can be accessed with the unique identifier GSE30601."], "basis": "The paper names GSE30601 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": ["cox_ph"], "assets": ["source:pmid:36779430", "data:geo:GSE30601"], "steps": ["run_analysis"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": [], "capabilities": [], "data_gaps": ["data:geo:GSE30601"]}, "gaps": []}, "review": {"state": "partial", "machine_critique": {"kind": "extraction_review", "reader": {"model": "MiniMax-M3.1-Flash", "provider": "minimax", "provider_family": "minimax", "read_at": "2026-11-01T18:27:16Z", "independence": "single_reader"}, "evidence_check": {"method": "verbatim", "availability": "verbatim"}, "calls": 5, "usage": {"prompt_tokens": 13676, "completion_tokens": 1621, "cached_tokens": 264}}, "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:36779430", "type": "paper", "label": "pmid:36779430"}, {"id": "source:pmid:36779430", "type": "manuscript", "label": "PMID 36779430 full text"}, {"id": "data:geo:GSE30601", "type": "dataset", "label": "GEO GSE30601"}], "activities": [{"id": "activity-reading-sci-2026-11-01", "type": "classification", "used": [{"id": "pmid:36779430", "type": "paper", "label": "pmid:36779430"}, {"id": "source:pmid:36779430", "type": "manuscript", "label": "PMID 36779430 full text"}, {"id": "data:geo:GSE30601", "type": "dataset", "label": "GEO GSE30601"}], "agent": "agent:corpus-assay-engine"}, {"id": "activity-read-36779430", "type": "extraction", "used": ["source:pmid:36779430"], "agent": "agent:minimax-reader"}, {"id": "activity-resolve-GSE30601", "type": "accession_resolution", "used": ["data:geo:GSE30601"], "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:36779430", "parent_hash": "sha256:4ab10956d2eadb8ee1d0dd59ccc655175e3911dc3ddcfa6b22b0e037c1265d09", "fork_of": null, "generation": 2}, "gaps": ["asset:data:geo:GSE30601 licence not read", "asset:data:geo:GSE30601 not downloaded", "execution.not_run", "method.reader_named_unmapped", "review.human_critique", "review.independent_reader", "review.machine_critique"], "assays": [{"role": "analysis", "definition_id": "survival_analysis", "definition_version": "f2132e96ae6c7a4e", "definition_digest": null, "required_operations": ["cox_ph"], "support_state": "supported", "source_hashes": {"dsl_catalogue": "f2132e96ae6c7a4e"}, "confidence": 0.7, "evidence": ["Cox regression"], "order_basis": "analysis_only", "id": "assay:1"}]}