Identification and verification of diagnostic biomarkers in recurrent pregnancy loss via machine learning algorithm and WGCNA.

{"sci_version": "1.2", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:37691920", "kind": "science-record", "revision": 4, "content_hash": "sha256:e38388bf592d9234a1b5695ea0cbb8b7c44e2f14844b0d05fea15f84066f47c7", "created_at": "2026-11-01T19:02:30Z", "updated_at": "2026-11-01T22:10:15Z", "generator": {"name": "corpus-assay-engine", "run_id": "reading-sci-2026-11-01"}, "paper": {"state": "complete", "title": "Identification and verification of diagnostic biomarkers in recurrent pregnancy loss via machine learning algorithm and WGCNA.", "doi": "10.3389/fimmu.2023.1241816", "pmid": "37691920", "authors": ["Changqiang Wei", "Yiyun Wei", "Jinlian Cheng", "Xuemei Tan", "Zhuolin Zhou", "Shanshan Lin", "Lihong Pang"], "journal": "Frontiers in immunology", "year": 2023, "abstract": "Background: Recurrent pregnancy loss defined as the occurrence of two or more pregnancy losses before 20-24 weeks of gestation, is a prevalent and significant pathological condition that impacts human reproductive health. However, the underlying mechanism of RPL remains unclear. This study aimed to investigate the biomarkers and molecular mechanisms associated with RPL and explore novel treatment strategies for clinical applications.\nMethods: The GEO database was utilized to retrieve the RPL gene expression profile GSE165004. This profile underwent differential expression analysis, WGCNA, functional enrichment, and subsequent analysis of RPL gene expression using LASSO regression, SVM-RFE, and RandomForest algorithms for hub gene screening. ANN model were constructed to assess the performance of hub genes in the dataset. The expression of hub genes in both the RPL and control group samples was validated using RT-qPCR. The immune cell infiltration level of RPL was assessed using CIBERSORT. Additionally, pan-cancer analysis was conducted using Sangerbox, and small-molecule drug screening was performed using CMap.\nResults: A total of 352 DEGs were identified, including 198 up-regulated genes and 154 down-regulated genes. Enrichment analysis indicated that the DEGs were primarily associated with Fc gamma R-mediated phagocytosis, the Fc epsilon RI signaling pathway, and various metabolism-related pathways. The turquoise module, which showed the highest relevance to clinical symptoms based on WGCNA results, contained 104 DEGs. Three hub genes, WBP11, ACTR2, and NCSTN, were identified using machine learning algorithms. ROC curves demonstrated a strong diagnostic value when the three hub genes were combined. RT-qPCR confirmed the low expression of WBP11 and ACTR2 in RPL, whereas NCSTN exhibited high expression. The immune cell infiltration analysis results indicated an imbalance of macrophages in RPL. Meanwhile, these three hub genes exhibited aberrant expression in multiple malignancies and were associated with a poor prognosis. Furthermore, we identified several small-molecule drugs.\nConclusion: This study identifies and validates hub genes in RPL, which may lead to significant advancements in understanding the molecular mechanisms and treatment strategies for this condition.", "claims": [{"id": "claim-method", "statement": "The discriminative ability of hub genes between experimental and control groups, as well as the accuracy of ANN model in the test and validation sets, were evaluated using receiver operating characteristic (ROC) curves.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:37691920", "locator": "sections[8] \"The artificial neural network prognostic model\" paragraph 2", "quote": "The discriminative ability of hub genes between experimental and control groups, as well as the accuracy of ANN model in the test and validation sets, were evaluated using receiver operating characteristic (ROC) curves."}], "state": "complete"}, {"id": "claim-data-access", "statement": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git.", "kind": "methodological", "target": null, "evidence": [{"asset": "source:pmid:37691920", "locator": "sections[25] \"Data availability statement\" paragraph 2", "quote": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git."}], "state": "complete"}, {"id": "claim-target-result", "statement": "The ROC results showed that the ANN model could effectively discriminate between RPL and controls, with an AUC value of 1.00 for the ROC curves for the test set GSE165004 and an AUC value of 0.74 for the validation set GSE183555 ( Figure 5D ).", "kind": "qualitative", "target": null, "evidence": [{"asset": "source:pmid:37691920", "locator": "sections[18] \"Diagnostic efficacy of hub biomarkers and ANN model\" paragraph 4", "quote": "The ROC results showed that the ANN model could effectively discriminate between RPL and controls, with an AUC value of 1.00 for the ROC curves for the test set GSE165004 and an AUC value of 0.74 for the validation set GSE183555 ( Figure 5D )."}], "state": "complete"}], "gaps": []}, "experiment": {"state": "partial", "objective": "The ROC results showed that the ANN model could effectively discriminate between RPL and controls, with an AUC value of 1.00 for the ROC curves for the test set GSE165004 and an AUC value of 0.74 for the validation set GSE183555 ( Figure 5D ).", "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:37691920"], "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:GSE165004"], "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:group_comparison_wilcoxon", "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:GSE165004"], "outputs": ["the replicated result and its comparison"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:run_analysis blocked on method:group_comparison_wilcoxon"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:37691920", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.3389/fimmu.2023.1241816", "sha256": "sha256:9b4b21fcbd5427bc91f719aba16678775d1bef73713bd4c7d84a6d4dfcf3f007", "bytes": 54049, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.3389/fimmu.2023.1241816", "version": null, "note": "Full article fetched from PMC; the checksum covers the rendered section body, not the original XML bytes.", "structure": {"pmcid": "PMC10485775", "source": "europe_pmc_fulltext", "coverage": "whole_body", "text_hash_scope": "rendered_section_body"}}, {"id": "data:geo:GSE165004", "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=GSE165004", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE165004", "version": null, "note": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git.", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git.", "access_statement_kind": "data_availability_statement", "sample_accessions": ["GSM5024320", "GSM5024321", "GSM5024322", "GSM5024323", "GSM5024324", "GSM5024325", "GSM5024326", "GSM5024327", "GSM5024328", "GSM5024329", "GSM5024330", "GSM5024331", "GSM5024332", "GSM5024333", "GSM5024334", "GSM5024335", "GSM5024336", "GSM5024337", "GSM5024338", "GSM5024339", "GSM5024340", "GSM5024341", "GSM5024342", "GSM5024343", "GSM5024344", "GSM5024345", "GSM5024346", "GSM5024347", "GSM5024348", "GSM5024349", "GSM5024350", "GSM5024351", "GSM5024352", "GSM5024353", "GSM5024354", "GSM5024355", "GSM5024356", "GSM5024357", "GSM5024358", "GSM5024359", "GSM5024360", "GSM5024361", "GSM5024362", "GSM5024363", "GSM5024364", "GSM5024365", "GSM5024366", "GSM5024367", "GSM5024368", "GSM5024369", "GSM5024370", "GSM5024371", "GSM5024372", "GSM5024373", "GSM5024374", "GSM5024375", "GSM5024376", "GSM5024377", "GSM5024378", "GSM5024379", "GSM5024380", "GSM5024381", "GSM5024382", "GSM5024383", "GSM5024384", "GSM5024385", "GSM5024386", "GSM5024387", "GSM5024388", "GSM5024389", "GSM5024390", "GSM5024391"], "sample_count": 72, "group_sizes": null, "group_sizes_basis": null, "platform": "16699", "taxon": "Homo sapiens", "title": "Endometrial Tissue RNA expression in recurrent pregnancy losses and unexplained infertility vs. conrol", "resolution_checked_at": "2026-11-01T18:06:43Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}, {"id": "data:geo:GSE183555", "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=GSE183555", "sha256": null, "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "GSE183555", "version": null, "note": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git.", "structure": {"repository": "GEO", "access_route": "GEO accession URL", "access_statement_verbatim": "The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git.", "access_statement_kind": "data_availability_statement", "sample_accessions": ["GSM5591871", "GSM5591872", "GSM5591873", "GSM5591874", "GSM5591875", "GSM5591876", "GSM5591877", "GSM5591878", "GSM5591879", "GSM5591880"], "sample_count": 10, "group_sizes": null, "group_sizes_basis": null, "platform": "21697", "taxon": "Homo sapiens", "title": "RNA sequencing of endometrial glands during the implantation window in women with recurrent pregnancy loss", "resolution_checked_at": "2026-11-01T19:09:46Z", "resolution_check": "identifier_resolution_only_no_download_no_licence_read", "downloaded": false, "licence_read": false}}], "gaps": ["asset:data:geo:GSE165004 not downloaded", "asset:data:geo:GSE165004 licence not read", "asset:data:geo:GSE183555 not downloaded", "asset:data:geo:GSE183555 licence not read"]}, "analysis": {"state": "partial", "methods": [{"id": "method-1", "family": "group_comparison", "specific": "wilcoxon", "label": "Group comparison", "dsl_operations": [], "missing_operations": ["group_comparison_wilcoxon"], "expressible": false, "parameters": {}, "data_shapes": ["two_or_more_groups_continuous"], "evidence": {"score": 0.8, "band": "high", "quote": "Wilcoxon"}}, {"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": "SVM"}}, {"id": "method-3", "family": null, "specific": null, "label": "receiver operating characteristic (ROC) curves", "dsl_operations": [], "missing_operations": [], "expressible": null, "parameters": {}, "data_shapes": [], "evidence": {"score": 1.0, "band": "verbatim", "quote": "The discriminative ability of hub genes between experimental and control groups, as well as the accuracy of ANN model in the test and validation sets, were evaluated using receiver operating characteristic (ROC) curves."}}], "targets": [], "gaps": ["method.reader_named_unmapped", "method.group_comparison_wilcoxon"]}, "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": ["group_comparison_wilcoxon"], "basis": "The paper names the method it used and the corpus's method extractor maps it onto the DSL catalogue: wilcoxon."}, "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": ["GSE165004", "GSE183555"], "resolved_accessions": ["GSE165004", "GSE183555"], "statements": ["The codes presented in the study are publicly available. This data can be found here: https://github.com/Weihouye/bioinformatics-analysis.git."], "basis": "The paper names GSE165004, GSE183555 in Gene Expression Omnibus (GEO) database and the accession resolves in the repository. No dataset was downloaded and no licence was read."}}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:37691920", "data:geo:GSE165004"], "steps": []}, "must_build": {"dsl_operations": [{"operation": "group_comparison_wilcoxon", "unlocks_papers": null, "priority": null}], "converters": [], "readers": [], "capabilities": [], "data_gaps": ["data:geo:GSE165004"]}, "gaps": []}, "review": {"state": "partial", "machine_critique": {"kind": "extraction_review", "reader": {"model": "MiniMax-M3.1-Flash", "provider": "minimax", "provider_family": "minimax", "read_at": "2026-11-01T18:52:57Z", "independence": "single_reader"}, "evidence_check": {"method": "verbatim", "availability": "verbatim"}, "calls": 5, "usage": {"prompt_tokens": 15415, "completion_tokens": 1367, "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:37691920", "type": "paper", "label": "pmid:37691920"}, {"id": "source:pmid:37691920", "type": "manuscript", "label": "PMID 37691920 full text"}, {"id": "data:geo:GSE165004", "type": "dataset", "label": "GEO GSE165004"}], "activities": [{"id": "activity-reading-sci-2026-11-01", "type": "classification", "used": [{"id": "pmid:37691920", "type": "paper", "label": "pmid:37691920"}, {"id": "source:pmid:37691920", "type": "manuscript", "label": "PMID 37691920 full text"}, {"id": "data:geo:GSE165004", "type": "dataset", "label": "GEO GSE165004"}], "agent": "agent:corpus-assay-engine"}, {"id": "activity-read-37691920", "type": "extraction", "used": ["source:pmid:37691920"], "agent": "agent:minimax-reader"}, {"id": "activity-resolve-GSE165004", "type": "accession_resolution", "used": ["data:geo:GSE165004"], "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:37691920", "parent_hash": "sha256:c4b0a3f401386fbabda122fa35343c7d256865508d1871d2c22b195a8f9f60c6", "fork_of": null, "generation": 3}, "gaps": ["asset:data:geo:GSE165004 licence not read", "asset:data:geo:GSE165004 not downloaded", "asset:data:geo:GSE183555 licence not read", "asset:data:geo:GSE183555 not downloaded", "execution.not_run", "method.group_comparison_wilcoxon", "method.reader_named_unmapped", "review.human_critique", "review.independent_reader", "review.machine_critique", "step:run_analysis blocked on method:group_comparison_wilcoxon"], "assays": [{"role": "data_generation", "definition_id": "qpcr", "definition_version": "99e1195c7d6f", "definition_digest": null, "required_operations": [], "support_state": "partial", "source_hashes": {"assay_lexicon": "99e1195c7d6f"}, "confidence": 0.99, "evidence": ["nd control group samples was validated using RT-qPCR. The immune cell infiltration level of RPL w", "alue when the three hub genes were combined. RT-qPCR confirmed the low expression of WBP11 and AC", "control group samples was validated using RT-qPCR. The immune cell infiltration level of RPL w", "e when the three hub genes were combined. RT-qPCR confirmed the low expression of WBP11 and AC"], "order_basis": "paper_text_order:1005", "id": "assay:1"}, {"role": "analysis", "definition_id": "group_comparison", "definition_version": "f2132e96ae6c7a4e", "definition_digest": null, "required_operations": [], "support_state": "partial", "source_hashes": {"dsl_catalogue": "f2132e96ae6c7a4e"}, "confidence": 0.8, "evidence": ["Wilcoxon"], "order_basis": "consumes_earlier_output", "id": "assay:2"}]}