{"id":"https://openalex.org/W7225238993","doi":"https://doi.org/10.64898/2026.11.03.763968","title":"Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines","display_name":"Where single-cell foundation models help: a benchmark of annotation, perturbation and regulatory inference against simple baselines","publication_year":2026,"publication_date":"2026-11-04","ids":{"openalex":"https://openalex.org/W7225238993","doi":"https://doi.org/10.64898/2026.11.03.763968"},"language":"en","primary_location":{"id":"doi:10.64898/2026.11.03.763968","doi":"https://doi.org/10.64898/2026.11.03.763968","is_oa":false,"landing_page_url":"https://doi.org/10.64898/2026.11.03.763968","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":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"volume":null,"type":"preprint","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5059630614","display_name":"Francesca Anna Cupaioli","orcid":"https://orcid.org/0000-0001-7518-1772","observed_orcids":["https://orcid.org/0000-0001-7518-1772"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Francesca Anna Cupaioli","raw_affiliation_strings":["CNR-ITB"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"CNR-ITB","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5026770450","display_name":"Ivan Merelli","orcid":"https://orcid.org/0000-0003-3587-3680","observed_orcids":["https://orcid.org/0000-0003-3587-3680"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ivan Merelli","raw_affiliation_strings":["CNR-ITB"],"raw_orcid":"https://orcid.org/0000-0003-3587-3680","affiliations":[{"raw_affiliation_string":"CNR-ITB","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"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/T11289","display_name":"Single-cell and spatial transcriptomics","score":0.9994999766349792,"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"}},"topics":[{"id":"https://openalex.org/T11289","display_name":"Single-cell and spatial transcriptomics","score":0.9994999766349792,"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/T10621","display_name":"Gene Regulatory Network Analysis","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"}},{"id":"https://openalex.org/T10885","display_name":"Gene expression and cancer classification","score":0.0,"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":[],"concepts":[{"id":"https://openalex.org/C2775905019","wikidata":"https://www.wikidata.org/wiki/Q192572","display_name":"In silico","level":3,"score":0.6306999921798706},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5701000094413757},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5188000202178955},{"id":"https://openalex.org/C70721500","wikidata":"https://www.wikidata.org/wiki/Q177005","display_name":"Computational biology","level":1,"score":0.44830000400543213},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.424699991941452},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.40549999475479126},{"id":"https://openalex.org/C177918212","wikidata":"https://www.wikidata.org/wiki/Q803623","display_name":"Perturbation (astronomy)","level":2,"score":0.388700008392334},{"id":"https://openalex.org/C146849305","wikidata":"https://www.wikidata.org/wiki/Q370766","display_name":"Ground truth","level":2,"score":0.3610000014305115},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3481999933719635},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3377000093460083},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3303000032901764},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.32839998602867126},{"id":"https://openalex.org/C98108389","wikidata":"https://www.wikidata.org/wiki/Q412563","display_name":"CRISPR","level":3,"score":0.30640000104904175},{"id":"https://openalex.org/C166998942","wikidata":"https://www.wikidata.org/wiki/Q16969238","display_name":"Interactor","level":2,"score":0.28690001368522644},{"id":"https://openalex.org/C111030470","wikidata":"https://www.wikidata.org/wiki/Q1430460","display_name":"Curse of dimensionality","level":2,"score":0.2858000099658966},{"id":"https://openalex.org/C179518139","wikidata":"https://www.wikidata.org/wiki/Q5140297","display_name":"Coding (social sciences)","level":2,"score":0.2596000134944916},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.25529998540878296},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.11.03.763968","doi":"https://doi.org/10.64898/2026.11.03.763968","is_oa":false,"landing_page_url":"https://doi.org/10.64898/2026.11.03.763968","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":null,"license_id":null,"version":"acceptedVersion","is_accepted":true,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"}],"best_oa_location":null,"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Single-cell":[0],"foundation":[1],"models":[2,39,255],"(FMs)":[3],"are":[4],"proposed":[5],"as":[6,158,160,248],"general":[7],"tools":[8],"for":[9,103,225],"cell-type":[10],"annotation,":[11],"in":[12,108,117,126],"silico":[13,109],"perturbation":[14,38],"and":[15,32,42,54,59,69,94,97,134,177,194,197,206,218,243],"gene-regulatory-network":[16],"(GRN)":[17],"inference.":[18],"To":[19],"identify":[20],"where":[21,172,262],"they":[22],"add":[23],"value,":[24],"we":[25],"benchmarked":[26],"five":[27],"FMs":[28,73,263],"(scGPT,":[29],"Geneformer":[30,95,135],"V1":[31],"V2,":[33],"Nicheformer,":[34],"scFoundation),":[35],"CellOracle,":[36],"supervised":[37],"including":[40],"STATE,":[41],"task-matched":[43],"simple":[44],"baselines":[45],"on":[46,259],"the":[47,104,112,127,173,183,190,222,238,244,249,260],"same":[48],"data,":[49],"from":[50,85,214],"K562":[51],"Perturb-seq":[52],"screens":[53],"primary":[55],"cells":[56,157],"to":[57,80,87,121,187],"healthy-donor":[58],"isogenic":[60,245],"leukodystrophy":[61],"brain":[62],"organoids,":[63,164,200],"a":[64,152],"public":[65],"organoid":[66],"CRISPR":[67],"screen":[68],"lung":[70],"cancer":[71],"biopsies.":[72],"were":[74,168],"most":[75],"useful":[76],"when":[77],"labels":[78],"had":[79],"be":[81,257],"transferred":[82],"across":[83],"species:":[84],"human":[86],"mouse":[88],"lung,":[89],"Nicheformer":[90],"reached":[91],"0.98":[92],"accuracy":[93,231],"V2":[96,136],"scFoundation":[98],"about":[99],"0.90,":[100],"against":[101],"0.68":[102],"best":[105],"baseline.":[106],"Zero-shot":[107],"perturbations":[110,124],"captured":[111],"direction":[113,129],"of":[114,123,189],"real":[115],"responses":[116],"K562,":[118],"with":[119,201],"up":[120],"75%":[122],"shifted":[125],"expected":[128],"(scFoundation,":[130],"p":[131],"<":[132],"1e-8),":[133],"networks":[137],"recovered":[138],"patient-level":[139],"tumour":[140],"differential":[141],"expression":[142],"better":[143],"than":[144],"co-expression":[145],"(fold":[146],"enrichment":[147],"2.9":[148],"versus":[149],"2.0).":[150],"Within":[151],"species,":[153],"FM":[154],"embeddings":[155],"annotated":[156],"accurately":[159],"principal":[161],"components.":[162],"In":[163],"predicted":[165],"knockout":[166,234],"effects":[167],"largely":[169],"explained":[170],"by":[171],"target":[174],"is":[175],"expressed,":[176],"an":[178],"expression-matched":[179],"control":[180],"test":[181],"isolated":[182],"gene-specific":[184],"part,":[185],"12":[186],"35%":[188],"effects.":[191],"Isogenic":[192],"Canavan":[193],"Krabbe":[195],"organoids":[196],"4H":[198],"patient":[199],"one":[202],"library":[203],"per":[204],"line":[205],"time":[207],"point,":[208],"did":[209],"not":[210,233],"yet":[211],"separate":[212],"methods":[213],"differences":[215],"between":[216],"clones":[217],"donors,":[219],"which":[220],"defines":[221],"replication":[223],"needed":[224],"future":[226],"validation.":[227],"Fine-tuning":[228],"changed":[229],"classification":[230],"but":[232],"rankings.":[235],"We":[236],"release":[237],"benchmark,":[239],"baselines,":[240],"pipeline":[241],"checks":[242],"reference":[246],"task":[247],"fmaudit":[250],"package,":[251],"so":[252],"that":[253],"new":[254],"can":[256],"evaluated":[258],"tasks":[261],"already":[264],"help.":[265]},"counts_by_year":[],"updated_date":"2026-11-05T12:25:35.769452","created_date":"2026-11-05T00:00:00"}