Kolmogorov-Arnold Network Model Integrated with Hypoxia Risk for Predicting PD-L1 Inhibitor Responses in Hepatocellular Carcinoma.

{"sci_version": "1.0", "schema": "https://corpus.bntrg.co.uk/sci/1.0/sci.schema.json", "id": "pmid:40150786", "kind": "science-record", "revision": 2, "content_hash": "sha256:f6753032c9938938497288a1b9fb35b36690a95c7d0ac2909e6d5c2f6e87a98d", "created_at": "2026-11-01T10:25:50Z", "updated_at": "2026-11-01T14:01:38Z", "generator": {"name": "corpus-assay-engine", "converter": "sep_to_sci_v1", "source_package_schema": "urn:bntrg:sep:1.0.0"}, "paper": {"state": "partial", "title": "Kolmogorov-Arnold Network Model Integrated with Hypoxia Risk for Predicting PD-L1 Inhibitor Responses in Hepatocellular Carcinoma.", "doi": "10.3390/bioengineering12030322", "pmid": "40150786", "authors": null, "journal": "Bioengineering (Basel, Switzerland)", "year": 2025, "abstract": null, "claims": [{"id": "claim-1", "statement": "the headline numerical result the paper states for its main finding", "kind": "quantitative", "state": "complete", "target": {"value": 0.725, "unit": "test accuracy", "tolerance": null, "source": "chars 1363-1444 of source sha256 09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09"}, "evidence": [{"asset": "source:pmid:40150786", "locator": "chars 1363-1444 of source sha256 09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09", "quote": "The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725."}]}], "gaps": ["abstract", "authors"]}, "experiment": {"state": "partial", "objective": "the headline numerical result the paper states for its main finding", "steps": [{"id": "read_published_estimate", "order": 1, "name": "Read the published estimate and its interval from the source", "kind": "other", "state": "complete", "blocked_on": null, "description": "The number a replication must land on, and the sentence it came from.", "duration_hours": null, "parameters": {}, "inputs": ["source:pmid:40150786"], "outputs": ["the published estimate, its unit and its interval"], "instruments": [], "reagents": []}, {"id": "obtain_source_data", "order": 2, "name": "Obtain the source data behind the published result", "kind": "data-preparation", "state": "blocked", "blocked_on": "data:not_downloaded", "description": "The published result cannot be recomputed without the data the paper analysed.", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the source data with its provenance"], "instruments": [], "reagents": []}, {"id": "run_analysis", "order": 3, "name": "Run the analysis programme", "kind": "analysis", "state": "blocked", "blocked_on": "no owned computational capability is bound to this experiment", "description": "Re-run the paper's analysis on the source data and produce the registered result. | SEP input not carried into the asset list: materials:the source data behind the published result in pmid:40150786", "duration_hours": null, "parameters": {}, "inputs": [], "outputs": ["the registered result"], "instruments": [], "reagents": []}], "materials": [], "instruments": [], "reagents": [], "conditions": {}, "gaps": ["step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40150786' does not resolve to a declared asset"]}, "data": {"state": "partial", "assets": [{"id": "source:pmid:40150786", "role": "manuscript", "format": "xml", "media_type": "application/xml", "uri": "https://doi.org/10.3390/bioengineering12030322", "sha256": "sha256:09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09", "bytes": null, "access": "open", "state": "resolved", "reader": null, "reader_status": "reader_required", "identifier": "10.3390/bioengineering12030322", "version": null, "note": "Source read by the package builder: full_text.", "structure": {"access_state": "full_text"}}, {"id": "asset:table:pmid:40150786:1", "role": "table", "format": "jats-table", "media_type": "application/vnd.jats.table+json", "uri": null, "sha256": "sha256:dbd8c7eb5136de49cd684003b94e0d23ec20541e6ca0a25ea9996ce38fe5c226", "bytes": 1089, "access": "open", "state": "verified", "reader": "read_table", "reader_status": "available", "identifier": "section[19]/table-wrap[1]#bioengineering-12-00322-t001", "version": null, "note": "Cox regression analysis for the hypoxia risk scoring model related to drug response prediction.", "structure": {"pmcid": "PMC11939538", "section_index": 19, "section_title": "3.2. Hypoxia Scoring Model", "row_count": 11, "column_count": 8}, "table": {"label": "Table 1", "caption": "Cox regression analysis for the hypoxia risk scoring model related to drug response prediction.", "rows": [["Gene", "log HR", "log HR SE", "HR", "t", "p", "95% CI Lower", "95%CI Upper"], ["PHLDA2", "0.1497", "0.0753", "1.1615", "1.9887", "0.0467", "1.0022", "1.3462"], ["DLGAP5", "0.0757", "0.2695", "1.0787", "0.281", "0.7787", "0.636", "1.8293"], ["N4BP2L1", "−0.2318", "0.1636", "0.7931", "−1.4163", "0.1567", "0.5755", "1.093"], ["CENPA", "0.099", "0.227", "1.104", "0.4361", "0.6628", "0.7076", "1.7226"], ["UPB1", "−0.0584", "0.0725", "0.9433", "−0.8061", "0.4202", "0.8184", "1.0872"], ["CABYR", "0.1509", "0.0752", "1.1629", "2.0063", "0.0448", "1.0035", "1.3476"], ["AFM", "−0.0022", "0.0609", "0.9978", "−0.0368", "0.9707", "0.8856", "1.1242"], ["HMMR", "0.3139", "0.1889", "1.3687", "1.6618", "0.0965", "0.9452", "1.982"], ["KIF20A", "0.0587", "0.2379", "1.0604", "0.2465", "0.8053", "0.6652", "1.6904"], ["PMAIP1", "−0.1203", "0.1857", "0.8866", "−0.648", "0.517", "0.6162", "1.2758"]], "row_count": 11, "column_count": 8, "numeric_cells": 73, "header_rows": 1, "locator": "section[19]/table-wrap[1]#bioengineering-12-00322-t001"}}, {"id": "asset:table:pmid:40150786:2", "role": "table", "format": "jats-table", "media_type": "application/vnd.jats.table+json", "uri": null, "sha256": "sha256:22d12d8e87a9931cf3f1fca2306f6ba2a686f616e544ee3979dfe9b4f01503f5", "bytes": 647, "access": "open", "state": "verified", "reader": "read_table", "reader_status": "available", "identifier": "section[20]/table-wrap[1]#bioengineering-12-00322-t002", "version": null, "note": "Hyperparameters and optimization settings for KAN training.", "structure": {"pmcid": "PMC11939538", "section_index": 20, "section_title": "3.3. Feature Selection and KAN Model", "row_count": 9, "column_count": 2}, "table": {"label": "Table 2", "caption": "Hyperparameters and optimization settings for KAN training.", "rows": [["Parameter", "Value"], ["Number of training processes (batches)", "6"], ["Number of grid intervals", "3–8"], ["Optimizer", "L-BFGS algorithm"], ["Learning rate", "1"], ["Maximum number of iterations per optimization process", "20"], ["Maximal number of function evaluations per optimization process", "25"], ["Termination tolerance on first-order optimality", "1 × 10−7"], ["Termination tolerance on function", "1 × 10−9"]], "row_count": 9, "column_count": 2, "numeric_cells": 7, "header_rows": 1, "locator": "section[20]/table-wrap[1]#bioengineering-12-00322-t002"}}], "gaps": ["asset:source:pmid:40150786 not verified"]}, "analysis": {"state": "absent", "methods": [], "targets": [], "gaps": ["method", "targets"]}, "capability": {"state": "absent", "requirements": [], "gaps": ["capability"]}, "readiness": {"state": "partial", "tier": null, "blocking_axis": null, "blocked_on": null, "axes": {}, "can_run": {"dsl_operations": [], "assets": ["source:pmid:40150786", "asset:table:pmid:40150786:1", "asset:table:pmid:40150786:2"], "steps": ["read_published_estimate"]}, "must_build": {"dsl_operations": [], "converters": [], "readers": ["xml"], "capabilities": [], "data_gaps": ["source:pmid:40150786"]}, "gaps": ["corpus_record_absent"]}, "review": {"state": "partial", "machine_critique": {"status": "blocked", "providers": [{"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "full-text-reading-v1", "prompt_hash": "d1f623687ee85d23ee2f76abab232d4358d9aaecc9601936a20c142d89f443a5", "source_hash": "09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09", "raw_response_sha256": "sha256:0be1d2781f587159de526076842edf4929bf46f5b76d0670048bae8c1040c49d", "raw_response_chars": 58291}], "target_providers": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "prompt_version": "replication-target-v2", "prompt_hash": "88911b9c9ce13943063c518dfbc62c72b82c8fda5d8c150f6a303197d09ec54c", "source_hash": "09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09", "raw_response": "{\"parent_value\":0.725,\"unit\":\"test accuracy\",\"evidence\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"target_result\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"number_stated\":true}", "reading": {"parent_value": 0.725, "unit": "test accuracy", "evidence": "The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.", "target_result": "The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.", "number_stated": true, "source_locator": "chars 1363-1444 of source sha256 09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09"}, "recorded_at": "2026-10-31T15:36:12.184684+00:00"}, {"provider": "minimax_alt", "model": "MiniMax-M3", "prompt_version": "replication-target-v2", "prompt_hash": "68df63bd0bf73f2d07654e291fdd2b4c8bdb8a416676ac9aae7d2d9085936b37", "source_hash": "09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09", "raw_response": "<think>We need solve extraction carefully. Need follow user output JSON only. Need identify \"single numerical result that a replication of the registered target has to reproduce\" \"headline number only: the value the paper itself states for the registered target, not a value you calculated, converted, averaged or reconstructed. The unit must be the unit the paper prints. ... REGISTERED TARGET: the headline numerical result the paper states for its main finding\". Source abstract several numbers: training test 290 patients, hypoxia model 10 genes, KAN 11 genes test accuracy 0.7, SVM integrated model test accuracy 0.725. Need choose headline main finding SVM integrating hypoxia and KAN models test accuracy 0.725 likely. Need parse. Title KAN Model Integrated with Hypoxia Risk for Predicting PD-L1 Inhibitor Responses in HCC. Main final integrated SVM achieves test accuracy 0.725. Target likely \"headline numerical result\" main finding. Evidence exact contiguous quotation \"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\" Unit \"test accuracy\" or \"accuracy\"? User unit must unit paper prints. Since accuracy 0.725 dimensionless, unit maybe \"test accuracy\". Need decide exact JSON parent_value 0.725, unit \"test accuracy\", evidence full sentence, target_result maybe \"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\" number_stated true. Or unit \"\" because no unit? User says unit must be unit paper prints. Paper prints \"test accuracy\" as metric. Often tasks extraction output units e.g. \"accuracy\". Need evaluate carefully, perhaps intended 0.725 and unit \"test accuracy\". We should not output another. Need consider if 290 is headline numerical result? No, main finding model accuracy. Abstract final main result test accuracy 0.725 vs KAN 0.7. Integrated SVM headline. Good.\n\nNeed think unit wording in JSON schema likely use \"test accuracy\" vs \"accuracy\". Could set unit: \"test accuracy\" directly printed. Value 0.725. Evidence exact clause, full sentence. target_result optional but required: say \"The integrated SVM model's test accuracy was 0.725.\" This paraphrases but target_result okay; requirement evidence exact only. Could target_result \"Test accuracy of the SVM integrating the hypoxia and KAN models: 0.725.\" Good. We should not include \"0.725\" number stated extra concerns? Fine.\n\nShould unit be \"\" as dimensionless? User unit must unit paper prints. They might expect unit \"accuracy\". We can choose \"test accuracy\". Need not invent \"%\". Clearly paper prints 0.725 not 72.5%, so unit not \"%\". \"test accuracy\" accurate metric. Output straightforward.\n\nWe have 30k tokens but no need. Could quote exact source \"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\" Full contiguous sentence. target_result mention integrated SVM test accuracy 0.725. No claim of computed. JSON only.\n\nFinal no markdown.</think>\n\n{\"parent_value\":0.725,\"unit\":\"test accuracy\",\"evidence\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"target_result\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"number_stated\":true}", "reading": {"parent_value": 0.725, "unit": "test accuracy", "evidence": "The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.", "target_result": "The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.", "number_stated": true, "source_locator": "chars 1363-1444 of source sha256 09c9fff11f795014b7c207b3f8802330ced1999811ab23ee9ea3e22835a50b09"}, "recorded_at": "2026-10-31T15:36:29.164705+00:00"}], "agreement": false, "disagreements": [], "agreement_basis": {"policy": "decisive_fields", "decisive_fields": ["work_type", "number_stated", "parent_value", "unit"]}, "blinded": true}, "human_critique": {"status": "not_run", "reviewer": null, "signed_at": null, "conflict_checked": false, "commentary": null}, "stages": [{"key": "target_materials", "status": "pending", "reason": null, "evidence": null}, {"key": "independent_readings", "status": "blocked", "reason": "The independent readings are incomplete or disagree; the source evidence must resolve the difference.", "evidence": {"provider_errors": [{"provider": "minimax", "model": "MiniMax-M3.1-Flash", "reason": "a repository quotation is absent from the supplied text", "raw_response": "{\"target_result\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"work_type\":\"analysis_only_for_target\",\"physical_methods\":[],\"analysis_methods\":[\"Differential expression analysis\",\"Enrichment analysis\",\"minimum Redundancy Maximum Relevance\",\"stepwise forward selection\",\"Synthetic Minority Over-sampling Technique\",\"Kolmogorov–Arnold Network\",\"Support Vector Machine\"],\"statistical_method\":\"test accuracy\",\"method_evidence\":\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"repositories\":[{\"repository\":\"European Genome-Phenome Archive\",\"accession\":\"EGAD00001008128\",\"data_description\":\"RNA sequencing data used for differential expression analysis and for assembling the training and test sets of 290 HCC patients.\",\"evidence\":\"Differential expression analysis (DEA) was performed on the EGAD00001008128, GSE41666, and GSE14520 datasets based on t-test q-values and fold change (FC) criteria. The training and test sets were assembled from the EGAD00001008128 dataset of 290 HCC patients\",\"publicly_downloadable\":true},{\"repository\":\"National Center for Biotechnology Information\",\"accession\":\"GSE233802\",\"data_description\":\"Dataset used to validate the hypoxia model.\",\"evidence\":\"Based on the HCC-Hypoxia Overlap (HHO) and immunotherapy response to hypoxia (IRH) genes selected by differential expression and enrichment analyses, a hypoxia model was built and validated on the TCGA-LIHC and GSE233802 datasets, respectively.\",\"publicly_downloadable\":true},{\"repository\":\"The Cancer Genome Atlas\",\"accession\":\"TCGA-LIHC\",\"data_description\":\"Dataset used to validate the hypoxia model.\",\"evidence\":\"Based on the HCC-Hypoxia Overlap (HHO) and immunotherapy response to hypoxia (IRH) genes selected by differential expression and enrichment analyses, a hypoxia model was built and validated on the TCGA-LIHC and GSE233802 datasets, respectively.\",\"publicly_downloadable\":true},{\"repository\":\"National Center for Biotechnology Information\",\"accession\":\"GSE41666\",\"data_description\":\"Microarray dataset used for differential expression analysis and identification of overlapping hypoxia-related genes.\",\"evidence\":\"Differential expression analysis (DEA) was performed on the EGAD00001008128, GSE41666, and GSE14520 datasets based on t-test q-values and fold change (FC) criteria.\",\"publicly_downloadable\":true},{\"repository\":\"National Center for Biotechnology Information\",\"accession\":\"GSE14520\",\"data_description\":\"Microarray dataset used for differential expression analysis and identification of overlapping hypoxia-related genes.\",\"evidence\":\"Differential expression analysis (DEA) was performed on the EGAD00001008128, GSE41666, and GSE14520 datasets based on t-test q-values and fold change (FC) criteria.\",\"publicly_downloadable\":true}],\"availability_statement\":\"The data presented in this study are openly available in European Genome-Phenome Archive (https://ega-archive.org/datasets/, accessed on 25 May 2024), National Center for Biotechnology Information database (https://www.ncbi.nlm.nih.gov/, accessed on 25 May 2024), and The Cancer Genome Atlas (https://www.cancerimagingarchive.net/, accessed on 25 May 2024).\",\"data_on_request\":false,\"source_claims\":[\"The KAN model with 11 genes achieved a test accuracy of 0.7.\",\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"The hypoxia model was constructed using 10 genes (IRH and HHO).\",\"The training and test sets were assembled from the EGAD00001008128 dataset of 290 HCC patients, and the response and non-response classes were balanced using the Synthetic Minority Over-sampling Technique.\",\"With the genes selected via the minimum Redundancy Maximum Relevance and stepwise forward methods, a Kolmogorov–Arnold Network (KAN) model was trained.\",\"The data presented in this study are openly available in European Genome-Phenome Archive (https://ega-archive.org/datasets/, accessed on 25 May 2024), National Center for Biotechnology Information database (https://www.ncbi.nlm.nih.gov/, accessed on 25 May 2024), and The Cancer Genome Atlas (https://www.cancerimagingarchive.net/, accessed on 25 May 2024).\"],\"missing_information\":[\"The paper does not report an inferential statistical test for the headline test-accuracy result; it reports the result using test accuracy.\",\"The supplied text does not provide accession-specific download instructions or explicitly state that each listed accession requires no application, embargo, or request; it states collectively that the data are openly available.\"],\"evidence_spans\":[\"The SVM integrating the hypoxia and KAN models achieved a test accuracy of 0.725.\",\"The KAN model with 11 genes achieved a test accuracy of 0.7.\",\"Based on the HCC-Hypoxia Overlap (HHO) and immunotherapy response to hypoxia (IRH) genes selected by differential expression and enrichment analyses, a hypoxia model was built and validated on the TCGA-LIHC and GSE233802 datasets, respectively.\",\"The training and test sets were assembled from the EGAD00001008128 dataset of 290 HCC patients, and the response and non-response classes were balanced using the Synthetic Minority Over-sampling Technique.\",\"Differential expression analysis (DEA) was performed on the EGAD00001008128, GSE41666, and GSE14520 datasets based on t-test q-values and fold change (FC) criteria.\",\"With the genes selected via the minimum Redundancy Maximum Relevance and stepwise forward methods, a Kolmogorov–Arnold Network (KAN) model was trained.\",\"The data presented in this study are openly available in European Genome-Phenome Archive (https://ega-archive.org/datasets/, accessed on 25 May 2024), National Center for Biotechnology Information database (https://www.ncbi.nlm.nih.gov/, accessed on 25 May 2024), and The Cancer Genome Atlas (https://www.cancerimagingarchive.net/, accessed on 25 May 2024).\"]}"}]}}, {"key": "source_resolution", "status": "pending", "reason": null, "evidence": null}, {"key": "protocol_reconstruction", "status": "pending", "reason": null, "evidence": null}, {"key": "control_execution", "status": "pending", "reason": null, "evidence": null}, {"key": "control_comparison", "status": "pending", "reason": null, "evidence": null}, {"key": "publication", "status": "pending", "reason": null, "evidence": null}], "gaps": ["review.human_critique", "review.machine_critique.blocked"]}, "execution": {"state": "absent", "status": "not_run", "mode": null, "runs": [], "gaps": ["execution.not_run"]}, "provenance": {"entities": [{"id": "package:e3f843db22382490bf8094bcaae0d57b3dcd18632b5112f6babd36741fb95654", "type": "sep_package", "label": "SEP v1 package for pmid:40150786"}], "activities": [{"id": "activity:conversion:pmid:40150786", "type": "conversion", "used": ["package:e3f843db22382490bf8094bcaae0d57b3dcd18632b5112f6babd36741fb95654"], "agent": "agent:corpus-assay-engine"}], "agents": [{"id": "agent:minimax", "type": "model", "label": "minimax"}, {"id": "agent:minimax_alt", "type": "model", "label": "minimax_alt"}, {"id": "agent:corpus-assay-engine", "type": "software", "label": "Corpus assay engine"}], "source_run_id": "run-20261023T122449232913", "dsl_catalogue_fingerprint": "343fb30756ba6ec3", "capability_table_fingerprint": "2cb3e42f047d2ecf", "snapshot_id": "ba34b49490d9bfaf"}, "lineage": {"parent_id": "pmid:40150786", "parent_hash": "sha256:b231628ac13e08ad3d4e90c04e1b34610b57887e7789c10bc6d37a8b7b065486", "fork_of": null, "generation": 2}, "gaps": ["abstract", "asset:source:pmid:40150786 not verified", "authors", "capability", "corpus_record_absent", "execution.not_run", "method", "review.human_critique", "review.machine_critique.blocked", "step:obtain_source_data blocked on data:not_downloaded", "step:run_analysis blocked on no owned computational capability is bound to this experiment", "step:run_analysis input 'materials:the source data behind the published result in pmid:40150786' does not resolve to a declared asset", "targets"]}