{"id":"https://openalex.org/W7219909797","doi":"https://doi.org/10.64898/2026.09.28.754066","title":"Mitigating Class Imbalance in Graph-Structured Data via Hierarchical Learning: Insights from Protein Binding Site Prediction","display_name":"Mitigating Class Imbalance in Graph-Structured Data via Hierarchical Learning: Insights from Protein Binding Site Prediction","publication_year":2026,"publication_date":"2026-10-04","ids":{"openalex":"https://openalex.org/W7219909797","doi":"https://doi.org/10.64898/2026.09.28.754066"},"language":"en","primary_location":{"id":"doi:10.64898/2026.09.28.754066","doi":"https://doi.org/10.64898/2026.09.28.754066","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.09.28.754066","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/10/04/2026.09.28.754066.full.pdf","source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://www.biorxiv.org/content/biorxiv/early/2026/10/04/2026.09.28.754066.full.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5083107866","display_name":"Vin\u00edcius de Almeida Paiva","orcid":"https://orcid.org/0000-0002-4411-1875","observed_orcids":["https://orcid.org/0000-0002-4411-1875"]},"institutions":[{"id":"https://openalex.org/I146165071","display_name":"Universidade Federal de Vi\u00e7osa","ror":"https://ror.org/0409dgb37","country_code":"BR","type":"education","lineage":["https://openalex.org/I146165071"]}],"countries":["BR"],"is_corresponding":true,"raw_author_name":"Vin\u00edcius Paiva","raw_affiliation_strings":["Universidade Federal de Vicosa"],"raw_orcid":"https://orcid.org/0000-0002-4411-1875","affiliations":[{"raw_affiliation_string":"Universidade Federal de Vicosa","institution_ids":["https://openalex.org/I146165071"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5156883799","display_name":"Leandro Marcolino","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210142044","display_name":"VinUniversity","ror":"https://ror.org/052dmdr17","country_code":"VN","type":"education","lineage":["https://openalex.org/I4210142044"]}],"countries":["VN"],"is_corresponding":false,"raw_author_name":"Leandro Marcolino","raw_affiliation_strings":["VinUniversity"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"VinUniversity","institution_ids":["https://openalex.org/I4210142044"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010423483","display_name":"Sandro Carvalho Izidoro","orcid":"https://orcid.org/0000-0001-5555-3321","observed_orcids":["https://orcid.org/0000-0001-5555-3321"]},"institutions":[{"id":"https://openalex.org/I8165072","display_name":"Universidade Federal de Itajub\u00e1","ror":"https://ror.org/00235nr42","country_code":"BR","type":"education","lineage":["https://openalex.org/I8165072"]}],"countries":["BR"],"is_corresponding":false,"raw_author_name":"Sandro Izidoro","raw_affiliation_strings":["Universidade Federal de Itajub"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Universidade Federal de Itajub","institution_ids":["https://openalex.org/I8165072"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021817826","display_name":"Sabrina de Azevedo Silveira","orcid":"https://orcid.org/0000-0002-4723-2349","observed_orcids":["https://orcid.org/0000-0002-4723-2349"]},"institutions":[{"id":"https://openalex.org/I146165071","display_name":"Universidade Federal de Vi\u00e7osa","ror":"https://ror.org/0409dgb37","country_code":"BR","type":"education","lineage":["https://openalex.org/I146165071"]}],"countries":["BR"],"is_corresponding":false,"raw_author_name":"Sabrina Silveira","raw_affiliation_strings":["Universidade Federal de Vicosa"],"raw_orcid":"https://orcid.org/0000-0002-4723-2349","affiliations":[{"raw_affiliation_string":"Universidade Federal de Vicosa","institution_ids":["https://openalex.org/I146165071"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5083107866"],"corresponding_institution_ids":["https://openalex.org/I146165071"],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":null,"last_page":null},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9520999789237976,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11273","display_name":"Advanced Graph Neural Networks","score":0.9520999789237976,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11652","display_name":"Imbalanced Data Classification Techniques","score":0.03739999979734421,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10211","display_name":"Computational Drug Discovery Methods","score":0.0017000000225380063,"subfield":{"id":"https://openalex.org/subfields/3002","display_name":"Drug Discovery"},"field":{"id":"https://openalex.org/fields/30","display_name":"Pharmacology, Toxicology and Pharmaceutics"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/class-imbalance","display_name":"class imbalance","score":0.9919999837875366},{"id":"https://openalex.org/keywords/graph-neural-networks","display_name":"graph neural networks","score":0.9490000009536743},{"id":"https://openalex.org/keywords/hierarchical-learning","display_name":"hierarchical learning","score":0.8820000290870667},{"id":"https://openalex.org/keywords/node-classification","display_name":"node classification","score":0.6759999990463257},{"id":"https://openalex.org/keywords/protein-ligand-binding-site-prediction","display_name":"protein-ligand binding site prediction","score":0.8600000143051147},{"id":"https://openalex.org/keywords/minority-class-detection","display_name":"minority class detection","score":0.5989999771118164},{"id":"https://openalex.org/keywords/clara","display_name":"Clara","score":0.824999988079071}],"concepts":[{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6527000069618225},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.625},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6039000153541565},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.6021999716758728},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5638999938964844},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4587000012397766},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.439300000667572},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.34380000829696655},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.33730000257492065},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.33379998803138733},{"id":"https://openalex.org/C45942800","wikidata":"https://www.wikidata.org/wiki/Q245652","display_name":"Ensemble learning","level":2,"score":0.31850001215934753},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.3109999895095825},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2937000095844269},{"id":"https://openalex.org/C2984842247","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep neural networks","level":3,"score":0.27309998869895935},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.2671000063419342},{"id":"https://openalex.org/C28225019","wikidata":"https://www.wikidata.org/wiki/Q4915005","display_name":"Biological network","level":2,"score":0.2669000029563904},{"id":"https://openalex.org/C201797286","wikidata":"https://www.wikidata.org/wiki/Q4914986","display_name":"Biological data","level":2,"score":0.25619998574256897}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.09.28.754066","doi":"https://doi.org/10.64898/2026.09.28.754066","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.09.28.754066","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/10/04/2026.09.28.754066.full.pdf","source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.64898/2026.09.28.754066","doi":"https://doi.org/10.64898/2026.09.28.754066","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.09.28.754066","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/10/04/2026.09.28.754066.full.pdf","source":{"id":"https://openalex.org/S4306402567","display_name":"bioRxiv (Cold Spring Harbor Laboratory)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":"https://openalex.org/I2750212522","host_organization_name":"Cold Spring Harbor Laboratory","host_organization_lineage":["https://openalex.org/I2750212522"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"sustainable_development_goals":[{"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10","score":0.5676000118255615}],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7219909797.pdf","grobid_xml":"https://content.openalex.org/works/W7219909797.grobid-xml"},"referenced_works_count":4,"referenced_works":["https://openalex.org/W4386271711","https://openalex.org/W4386566925","https://openalex.org/W4404261350","https://openalex.org/W4416011383"],"related_works":["https://openalex.org/W4416428259","https://openalex.org/W4396723207","https://openalex.org/W4410916958","https://openalex.org/W4405956625","https://openalex.org/W7127566114","https://openalex.org/W4403467909","https://openalex.org/W4220808225","https://openalex.org/W4388184025","https://openalex.org/W4221150256","https://openalex.org/W4312045632"],"abstract_inverted_index":{"Abstract":[0],"Learning":[1],"from":[2],"imbalanced":[3],"data":[4],"remains":[5],"a":[6,34,45,59],"major":[7],"challenge":[8],"for":[9],"graph":[10,87,125],"neural":[11],"networks":[12],"(GNNs),":[13],"as":[14],"minority":[15,55],"nodes":[16],"are":[17],"not":[18],"only":[19],"rare":[20],"but":[21],"also":[22],"structurally":[23],"marginalized":[24],"within":[25,63],"the":[26],"graph.":[27],"We":[28],"address":[29],"this":[30],"issue":[31],"with":[32,96],"CLARA,":[33],"hierarchical":[35],"learning":[36,79,126],"framework":[37],"that":[38,49],"decomposes":[39],"node":[40],"classification":[41],"into":[42],"two":[43],"stages:":[44],"coarse":[46],"subgraph-level":[47],"classifier":[48],"selects":[50],"regions":[51],"likely":[52],"to":[53],"contain":[54],"instances,":[56],"followed":[57],"by":[58],"fine-grained":[60],"node-level":[61],"predictor":[62],"these":[64],"regions.":[65],"This":[66],"design":[67],"improves":[68],"sensitivity":[69],"while":[70],"maintaining":[71],"scalability,":[72],"filtering":[73],"out":[74],"irrelevant":[75],"areas":[76],"and":[77,117,123],"focusing":[78],"on":[80,85],"topologically":[81],"meaningful":[82],"neighborhoods.":[83],"Experiments":[84],"benchmark":[86],"datasets":[88],"demonstrate":[89],"substantial":[90],"gains":[91],"over":[92],"established":[93],"imbalance-handling":[94],"methods,":[95],"CLARA":[97],"reaching":[98],"an":[99],"F1-score":[100],"of":[101],"88.3%.":[102],"The":[103],"same":[104],"strategy":[105],"achieves":[106],"significant":[107],"improvements":[108],"in":[109],"protein\u2013ligand":[110],"binding":[111],"site":[112],"prediction,":[113],"underscoring":[114],"its":[115],"broad":[116],"consistent":[118],"effectiveness":[119],"across":[120],"both":[121],"biological":[122],"general":[124],"tasks.":[127]},"counts_by_year":[],"updated_date":"2026-11-08T12:12:28.789561","created_date":"2026-10-05T00:00:00"}