{"id":"https://openalex.org/W7166706745","doi":"https://doi.org/10.64898/2026.06.29.735309","title":"Identifying and Addressing Systematic Data Leakage in Protein-Ligand Affinity Benchmarks","display_name":"Identifying and Addressing Systematic Data Leakage in Protein-Ligand Affinity Benchmarks","publication_year":2026,"publication_date":"2026-06-30","ids":{"openalex":"https://openalex.org/W7166706745","doi":"https://doi.org/10.64898/2026.06.29.735309"},"language":"en","primary_location":{"id":"doi:10.64898/2026.06.29.735309","doi":"https://doi.org/10.64898/2026.06.29.735309","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.06.29.735309","pdf_url":null,"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://doi.org/10.64898/2026.06.29.735309","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139554303","display_name":"Bj\u00f6rn Mattsson","orcid":"https://orcid.org/0009-0004-3995-065X","observed_orcids":["https://orcid.org/0009-0004-3995-065X"]},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Bj\u00f6rn Mattsson","raw_affiliation_strings":["Enlace Bio"],"raw_orcid":"https://orcid.org/0009-0004-3995-065X","affiliations":[{"raw_affiliation_string":"Enlace Bio","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5077218712","display_name":"W. Patrick Walters","orcid":"https://orcid.org/0000-0003-2860-7958","observed_orcids":["https://orcid.org/0000-0003-2860-7958"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"W. Patrick Walters","raw_affiliation_strings":["OpenADMET"],"raw_orcid":"https://orcid.org/0000-0003-2860-7958","affiliations":[{"raw_affiliation_string":"OpenADMET","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":["https://openalex.org/A5139554303"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":7,"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/T10211","display_name":"Computational Drug Discovery Methods","score":0.9990000128746033,"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"}},"topics":[{"id":"https://openalex.org/T10211","display_name":"Computational Drug Discovery Methods","score":0.9990000128746033,"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"}},{"id":"https://openalex.org/T10044","display_name":"Protein Structure and Dynamics","score":0.00019999999494757503,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}},{"id":"https://openalex.org/T12576","display_name":"vaccines and immunoinformatics approaches","score":9.999999747378752e-05,"subfield":{"id":"https://openalex.org/subfields/1312","display_name":"Molecular Biology"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/protein-ligand-binding-affinity-prediction","display_name":"protein-ligand binding affinity prediction","score":0.9390000104904175},{"id":"https://openalex.org/keywords/data-leakage","display_name":"data leakage","score":0.8920000195503235},{"id":"https://openalex.org/keywords/benchmark-design","display_name":"benchmark design","score":0.6830000281333923},{"id":"https://openalex.org/keywords/protein-sequence-identity","display_name":"protein sequence identity","score":0.5759999752044678},{"id":"https://openalex.org/keywords/boltz-2","display_name":"Boltz-2","score":0.9340000152587891},{"id":"https://openalex.org/keywords/free-energy-perturbation","display_name":"free energy perturbation","score":0.824999988079071},{"id":"https://openalex.org/keywords/chembl","display_name":"ChEMBL","score":0.6589999794960022},{"id":"https://openalex.org/keywords/model-generalization","display_name":"model generalization","score":0.6539999842643738},{"id":"https://openalex.org/keywords/structure-based-drug-design","display_name":"structure-based drug design","score":0.8920000195503235}],"concepts":[{"id":"https://openalex.org/C63222358","wikidata":"https://www.wikidata.org/wiki/Q6120337","display_name":"chEMBL","level":3,"score":0.6952000260353088},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6281999945640564},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.5231000185012817},{"id":"https://openalex.org/C2778738651","wikidata":"https://www.wikidata.org/wiki/Q16546687","display_name":"Novelty","level":2,"score":0.5199000239372253},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5166000127792358},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.5008999705314636},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4034000039100647},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.3671000003814697},{"id":"https://openalex.org/C2777042071","wikidata":"https://www.wikidata.org/wiki/Q6509304","display_name":"Leakage (economics)","level":2,"score":0.3571999967098236},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C95623464","wikidata":"https://www.wikidata.org/wiki/Q1096149","display_name":"Classifier (UML)","level":2,"score":0.3075999915599823},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C75608658","wikidata":"https://www.wikidata.org/wiki/Q44395","display_name":"Pascal (unit)","level":2,"score":0.29109999537467957},{"id":"https://openalex.org/C30038468","wikidata":"https://www.wikidata.org/wiki/Q4354775","display_name":"Memorization","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C111364199","wikidata":"https://www.wikidata.org/wiki/Q2915896","display_name":"Protein methods","level":4,"score":0.2605000138282776}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.06.29.735309","doi":"https://doi.org/10.64898/2026.06.29.735309","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.06.29.735309","pdf_url":null,"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:10.64898/2026.06.29.735309","doi":"https://doi.org/10.64898/2026.06.29.735309","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.06.29.735309","pdf_url":null,"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":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W1968319881","https://openalex.org/W1982131304","https://openalex.org/W1984052581","https://openalex.org/W2068385142","https://openalex.org/W2092285329","https://openalex.org/W2096864392","https://openalex.org/W2587598315","https://openalex.org/W2803094965","https://openalex.org/W2918239264","https://openalex.org/W2950954328","https://openalex.org/W2969325194","https://openalex.org/W4297914802","https://openalex.org/W4376131268","https://openalex.org/W4385576721","https://openalex.org/W4387639561","https://openalex.org/W4388455891","https://openalex.org/W4400795417","https://openalex.org/W4410138743","https://openalex.org/W4411174340","https://openalex.org/W4413942118","https://openalex.org/W4415384230","https://openalex.org/W7117321007","https://openalex.org/W7118191769","https://openalex.org/W7160647050"],"related_works":["https://openalex.org/W7214103012","https://openalex.org/W7118191769","https://openalex.org/W7115907569","https://openalex.org/W4417127631","https://openalex.org/W7126022351","https://openalex.org/W4313551409","https://openalex.org/W4394906324","https://openalex.org/W4405308288","https://openalex.org/W7163425806","https://openalex.org/W7137943277"],"abstract_inverted_index":{"ABSTRACT":[0],"Accurate":[1],"prediction":[2],"of":[3,27,47,62,121],"protein-ligand":[4],"binding":[5,117],"affinity":[6],"is":[7,93,228],"a":[8,24,59,77,160,253],"crucial":[9],"goal":[10],"in":[11,104,123,153,223],"structure-based":[12],"drug":[13],"discovery,":[14],"with":[15,108],"the":[16,124,149,174,219,225,235,263],"potential":[17],"to":[18,96,101,196,270],"significantly":[19],"shorten":[20],"development":[21],"timelines.":[22],"Recently,":[23],"new":[25,78],"wave":[26],"machine":[28],"learning":[29,210],"models":[30,198,244],"based":[31],"on":[32,173],"co-folding,":[33],"such":[34,132],"as":[35,143,145],"Boltz-2":[36],"and":[37,75,135,177,184,203],"IsoDDE,":[38],"has":[39],"demonstrated":[40],"performance":[41,172],"that":[42,46,66,80,88,137,159,192,262,272],"matches":[43],"or":[44],"exceeds":[45],"gold-standard":[48],"physics-based":[49],"methods":[50],"like":[51],"Free":[52],"Energy":[53],"Perturbation":[54],"(FEP).":[55],"This":[56],"paper":[57],"provides":[58],"critical":[60],"assessment":[61],"these":[63],"claims,":[64],"revealing":[65],"current":[67,193],"benchmarks":[68,154,179,194],"are":[69],"heavily":[70],"influenced":[71],"by":[72,90],"data":[73,84,98,202,227],"leakage,":[74],"proposes":[76],"benchmark":[79],"explicitly":[81],"controls":[82],"for":[83,140,199,256],"leakage.":[85],"We":[86,260],"demonstrate":[87],"splitting":[89],"protein-sequence":[91],"identity":[92,112],"inherently":[94],"insufficient":[95],"prevent":[97],"leakage":[99,138,206],"due":[100],"\u201ctarget":[102],"mirroring,\u201d":[103],"which":[105,164,224],"homologous":[106],"proteins":[107],"low":[109,144],"overall":[110],"sequence":[111],"still":[113],"exhibit":[114],"highly":[115],"correlated":[116],"profiles.":[118],"Our":[119,189],"meta-analysis":[120],"documents":[122],"ChEMBL":[125],"36":[126],"database":[127],"identifies":[128],"more":[129],"than":[130,208],"6,000":[131],"assay":[133],"pairs":[134],"finds":[136],"persists":[139],"sequence-identity":[141],"thresholds":[142],"0.2,":[146],"well":[147],"below":[148],"values":[150],"commonly":[151],"used":[152],"today.":[155],"Additionally,":[156],"we":[157,217],"show":[158],"ligand-only":[161,243],"baseline":[162,255],"model,":[163],"lacks":[165],"protein":[166],"structural":[167],"information,":[168],"achieves":[169],"surprisingly":[170],"high":[171],"FEP+":[175],"4":[176],"OpenFE":[178],"(":[180,248],"r":[181,185,249],"=":[182,186,250],"0.66":[183],"0.36,":[187],"respectively).":[188],"results":[190],"indicate":[191],"tend":[195],"reward":[197],"memorizing":[200],"training":[201],"exploiting":[204],"localized":[205],"rather":[207],"truly":[209],"biophysical":[211],"principles.":[212],"To":[213],"address":[214],"this":[215],"issue,":[216],"propose":[218],"Novelty-Tiered":[220],"Affinity":[221],"Benchmark,":[222],"test":[226],"partitioned":[229],"into":[230,276],"ligand":[231],"novelty":[232],"tiers.":[233],"In":[234],"most":[236],"challenging":[237],"tier":[238],"(Tanimoto":[239],"similarity":[240],"<":[241],"0.35),":[242],"perform":[245],"notably":[246],"worse":[247],"0.14),":[251],"offering":[252],"clear":[254],"evaluating":[257],"genuine":[258],"generalization.":[259],"argue":[261],"field":[264],"must":[265],"move":[266],"beyond":[267],"sequence-based":[268],"splits":[269],"ensure":[271],"AI-driven":[273],"discovery":[274],"translates":[275],"successful":[277],"prospective":[278],"laboratory":[279],"research.":[280]},"counts_by_year":[{"year":2026,"cited_by_count":7}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-07-01T00:00:00"}