{"id":"https://openalex.org/W7204997949","doi":"https://doi.org/10.64898/2026.08.31.747783","title":"Scaling recipes for single-cell RNA sequencing foundation models: when do scaling laws hold?","display_name":"Scaling recipes for single-cell RNA sequencing foundation models: when do scaling laws hold?","publication_year":2026,"publication_date":"2026-09-01","ids":{"openalex":"https://openalex.org/W7204997949","doi":"https://doi.org/10.64898/2026.08.31.747783"},"language":"en","primary_location":{"id":"doi:10.64898/2026.08.31.747783","doi":"https://doi.org/10.64898/2026.08.31.747783","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.31.747783","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/09/01/2026.08.31.747783.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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"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/2026/09/01/2026.08.31.747783.full.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5035464509","display_name":"Federico Borra","orcid":"https://orcid.org/0000-0003-1385-8877","observed_orcids":["https://orcid.org/0000-0003-1385-8877"]},"institutions":[{"id":"https://openalex.org/I4210114103","display_name":"IFOM","ror":"https://ror.org/02hcsa680","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210114103"]},{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Federico Borra","raw_affiliation_strings":["*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","institution_ids":["https://openalex.org/I4210114103"]},{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128404003","display_name":"Giacomo Cir\u00f2","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4210114103","display_name":"IFOM","ror":"https://ror.org/02hcsa680","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210114103"]},{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Giacomo Cir\u00f2","raw_affiliation_strings":["*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","institution_ids":["https://openalex.org/I4210114103"]},{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134456357","display_name":"Arianna Castellini","orcid":"https://orcid.org/0009-0006-7410-3231","observed_orcids":["https://orcid.org/0009-0006-7410-3231"]},"institutions":[{"id":"https://openalex.org/I4210114103","display_name":"IFOM","ror":"https://ror.org/02hcsa680","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210114103"]},{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Arianna Castellini","raw_affiliation_strings":["*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","institution_ids":["https://openalex.org/I4210114103"]},{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5147418997","display_name":"Giovanni Gatti","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Giovanni Gatti","raw_affiliation_strings":["Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056055676","display_name":"Andrea Tangherloni","orcid":"https://orcid.org/0000-0002-5856-4453","observed_orcids":["https://orcid.org/0000-0002-5856-4453"]},"institutions":[{"id":"https://openalex.org/I4210114103","display_name":"IFOM","ror":"https://ror.org/02hcsa680","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210114103"]},{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Andrea Tangherloni","raw_affiliation_strings":["*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","institution_ids":["https://openalex.org/I4210114103"]},{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5075865182","display_name":"Francesca Meteora Buffa","orcid":"https://orcid.org/0000-0003-0409-406X","observed_orcids":["https://orcid.org/0000-0003-0409-406X"]},"institutions":[{"id":"https://openalex.org/I4210114103","display_name":"IFOM","ror":"https://ror.org/02hcsa680","country_code":"IT","type":"facility","lineage":["https://openalex.org/I4210114103"]},{"id":"https://openalex.org/I71209653","display_name":"Bocconi University","ror":"https://ror.org/05crjpb27","country_code":"IT","type":"education","lineage":["https://openalex.org/I71209653"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Francesca M. Buffa","raw_affiliation_strings":["*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy"],"raw_orcid":"https://orcid.org/0000-0003-0409-406X","affiliations":[{"raw_affiliation_string":"*AI & Systems Biology Lab, IFOM, Via Adamello, 16, Milan, 20139, Lombardy, Italy","institution_ids":["https://openalex.org/I4210114103"]},{"raw_affiliation_string":"Computing Sciences Department, Bocconi University, Via Sarfatti, 25, Milan, 20136, Lombardy, Italy","institution_ids":["https://openalex.org/I71209653"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5075865182"],"corresponding_institution_ids":["https://openalex.org/I4210114103","https://openalex.org/I71209653"],"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/T11289","display_name":"Single-cell and spatial transcriptomics","score":0.9945999979972839,"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":0.9945999979972839,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.00039999998989515007,"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/T10885","display_name":"Gene expression and cancer classification","score":0.0003000000142492354,"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/scaling-laws","display_name":"scaling laws","score":0.847000002861023},{"id":"https://openalex.org/keywords/foundation-models","display_name":"foundation models","score":0.8510000109672546},{"id":"https://openalex.org/keywords/single-cell-rna-sequencing","display_name":"single-cell RNA sequencing","score":0.8849999904632568},{"id":"https://openalex.org/keywords/single-cell-transcriptomics","display_name":"single-cell transcriptomics","score":0.6769999861717224},{"id":"https://openalex.org/keywords/model-capacity","display_name":"model capacity","score":0.7820000052452087},{"id":"https://openalex.org/keywords/power-law-scaling","display_name":"power-law scaling","score":0.7080000042915344},{"id":"https://openalex.org/keywords/learning-rate","display_name":"learning rate","score":0.7760000228881836},{"id":"https://openalex.org/keywords/depth-to-width-ratio","display_name":"depth-to-width ratio","score":0.9919999837875366},{"id":"https://openalex.org/keywords/hyperparameter-optimization","display_name":"hyperparameter optimization","score":0.8370000123977661},{"id":"https://openalex.org/keywords/model-architecture","display_name":"model architecture","score":0.5080000162124634}],"concepts":[{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.7269999980926514},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.7129999995231628},{"id":"https://openalex.org/C2988430800","wikidata":"https://www.wikidata.org/wiki/Q428971","display_name":"Scaling law","level":3,"score":0.6237999796867371},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5856000185012817},{"id":"https://openalex.org/C8642999","wikidata":"https://www.wikidata.org/wiki/Q4171168","display_name":"Hyperparameter","level":2,"score":0.5496000051498413},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.484499990940094},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46209999918937683},{"id":"https://openalex.org/C87040749","wikidata":"https://www.wikidata.org/wiki/Q428971","display_name":"Power law","level":2,"score":0.39730000495910645},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.38989999890327454},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.38920000195503235},{"id":"https://openalex.org/C120936955","wikidata":"https://www.wikidata.org/wiki/Q2155640","display_name":"Empirical research","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C149782125","wikidata":"https://www.wikidata.org/wiki/Q160039","display_name":"Econometrics","level":1,"score":0.3249000012874603},{"id":"https://openalex.org/C133199616","wikidata":"https://www.wikidata.org/wiki/Q25386885","display_name":"Empirical modelling","level":2,"score":0.32350000739097595},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.31869998574256897},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.30889999866485596},{"id":"https://openalex.org/C163258240","wikidata":"https://www.wikidata.org/wiki/Q25342","display_name":"Power (physics)","level":2,"score":0.3037000000476837},{"id":"https://openalex.org/C2522767166","wikidata":"https://www.wikidata.org/wiki/Q2374463","display_name":"Data science","level":1,"score":0.30219998955726624},{"id":"https://openalex.org/C2780600066","wikidata":"https://www.wikidata.org/wiki/Q7239828","display_name":"Preferential attachment","level":3,"score":0.2888000011444092},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.28450000286102295},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.26980000734329224},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.26899999380111694},{"id":"https://openalex.org/C76969082","wikidata":"https://www.wikidata.org/wiki/Q486902","display_name":"Mathematical model","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C170122806","wikidata":"https://www.wikidata.org/wiki/Q1914828","display_name":"Linear scale","level":2,"score":0.26030001044273376}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.64898/2026.08.31.747783","doi":"https://doi.org/10.64898/2026.08.31.747783","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.31.747783","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/09/01/2026.08.31.747783.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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"}],"best_oa_location":{"id":"doi:10.64898/2026.08.31.747783","doi":"https://doi.org/10.64898/2026.08.31.747783","is_oa":true,"landing_page_url":"https://doi.org/10.64898/2026.08.31.747783","pdf_url":"https://www.biorxiv.org/content/biorxiv/early/2026/09/01/2026.08.31.747783.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-nc-nd","license_id":"https://openalex.org/licenses/cc-by-nc-nd","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"bioRxiv","raw_type":"posted-content"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204997949.pdf","grobid_xml":"https://content.openalex.org/works/W7204997949.grobid-xml"},"referenced_works_count":22,"referenced_works":["https://openalex.org/W1757909678","https://openalex.org/W2612519453","https://openalex.org/W2749581528","https://openalex.org/W2947629474","https://openalex.org/W2970971581","https://openalex.org/W3030163527","https://openalex.org/W3035965352","https://openalex.org/W4225598893","https://openalex.org/W4378838672","https://openalex.org/W4392168151","https://openalex.org/W4399387478","https://openalex.org/W4401314781","https://openalex.org/W4404793799","https://openalex.org/W4405966441","https://openalex.org/W4409566559","https://openalex.org/W4409624394","https://openalex.org/W4411757935","https://openalex.org/W4412929241","https://openalex.org/W6950179980","https://openalex.org/W7093313664","https://openalex.org/W7119591224","https://openalex.org/W7133225605"],"related_works":["https://openalex.org/W7164051593","https://openalex.org/W7171662732","https://openalex.org/W7130463522","https://openalex.org/W4399387478","https://openalex.org/W7163822972","https://openalex.org/W7141681014","https://openalex.org/W4405966441","https://openalex.org/W4415928258","https://openalex.org/W7169771264","https://openalex.org/W7163700209"],"abstract_inverted_index":{"Abstract":[0],"Deep":[1],"learning":[2,96],"models":[3,48,147],"exhibit":[4],"empirical":[5,91],"scaling":[6,44,112],"laws":[7],"whereby":[8],"performance":[9],"changes":[10],"predictably":[11],"with":[12,63],"model":[13,64,74,85,102],"size,":[14,16],"dataset":[15],"and":[17,31,66,78,89,98,104,133,137],"training":[18,67],"compute.":[19,107],"Although":[20],"these":[21,81],"relationships":[22,92],"are":[23],"well":[24],"established":[25],"in":[26,46],"domains":[27],"such":[28],"as":[29],"language":[30],"image":[32],"modelling,":[33],"their":[34],"applicability":[35],"to":[36,101,115],"biological":[37],"data":[38],"remains":[39],"unclear.":[40],"Here,":[41],"we":[42],"investigate":[43],"behaviour":[45],"foundation":[47,146],"trained":[49],"on":[50,73],"large":[51],"collections":[52],"of":[53,80,143],"single-cell":[54],"transcriptomes.":[55],"We":[56,87],"show":[57],"that":[58,111],"pre-training":[59],"loss":[60],"decreases":[61],"systematically":[62],"capacity":[65],"compute,":[68],"exhibiting":[69],"a":[70,122],"power-law":[71],"dependence":[72],"size.":[75],"The":[76],"strength":[77],"regularity":[79],"trends":[82],"differ":[83],"between":[84],"formulations.":[86],"identify":[88],"quantify":[90],"linking":[93],"the":[94,127,141],"optimal":[95],"rate":[97],"depth-to-width":[99],"ratio":[100],"size":[103],"depth":[105],"or":[106],"These":[108],"results":[109],"demonstrate":[110],"principles":[113],"extend":[114],"transcriptomic":[116],"modelling.":[117],"More":[118],"broadly,":[119],"they":[120],"provide":[121],"quantitative":[123],"framework":[124],"for":[125,148],"estimating":[126],"expected":[128],"returns":[129],"from":[130],"additional":[131],"resources":[132],"selecting":[134],"suitable":[135],"hyperparameters":[136],"architectures,":[138],"thereby":[139],"supporting":[140],"development":[142],"increasingly":[144],"capable":[145],"omics":[149],"data.":[150]},"counts_by_year":[],"updated_date":"2026-11-09T13:22:03.289127","created_date":"2026-09-02T00:00:00"}