{"id":{"repo_id":"msu","oai_identifier":"oai:d.lib.msu.edu:etd_53208"},"canonical_url":"https://search.ndltd.org/etd/msu/oai:d.lib.msu.edu:etd_53208","repository":{"repo_id":"msu","name":"Michigan State University","base_url":"https://d.lib.msu.edu/oai"},"display":{"title":"Towards Foundation Model for Single-Cell Biology","abstract":"Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2026","abstract_html":"Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2026","abstract_has_math":false,"creators":["Tang, Wenzhuo"],"institution":null,"degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Xie, Yuying","Cui, Yuehua","Hu, Guanyu","Tan, Pang-Ning"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2026,"date_issued":"2026","date_published":"2026","updated_at":"2026-11-18T18:59:44Z","subjects":["Statistics"],"languages":["eng"],"rights":["Attribution 4.0 International"],"rights_urls":[],"identifier_entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53208","local:Tang_grad.msu_0128D_22237"],"render_values":[{"text":"etd:53208","href":null,"code":true},{"text":"local:Tang_grad.msu_0128D_22237","href":null,"code":true}]}],"person_identifier_entries":[]},"links":{"outbound_url":"https://doi.org/doi:10.25335/7xbe-ts29","outbound_label":"DOI","outbound_source":"dc:identifier","record_url":"https://doi.org/doi:10.25335/7xbe-ts29","record_label":"Repository record","record_kind":"doi","record_source":"dc:identifier","download_url":null,"download_label":null,"download_source":null,"access_status":null,"embargo_until":null,"access_note":null},"source_record":{"url":"https://d.lib.msu.edu/oai?verb=GetRecord&metadataPrefix=oai_dc&identifier=oai%3Ad.lib.msu.edu%3Aetd_53208","prefix":"oai_dc"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Xie, Yuying","Cui, Yuehua","Hu, Guanyu","Tan, Pang-Ning"]},{"key":"dc:creator","label":"Author","values":["Tang, Wenzhuo"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2026"]},{"key":"dc:relation","label":"Dc Relation","values":["Electronic Theses & Dissertations"]},{"key":"dc:type","label":"Dc Type","values":["text","Electronic dissertations"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Statistics"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Attribution 4.0 International"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["etd:53208","local:Tang_grad.msu_0128D_22237","https://doi.org/doi:10.25335/7xbe-ts29"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2026","Single-cell and spatial profiling technologies have created new opportunities to study biologicalsystems at cellular resolution, generating large-scale datasets across diverse tissues, organisms, modalities, developmental stages, disease conditions, and perturbations. These data provide rich views of cellular states and biological context, but they are also high-dimensional, sparse, noisy, heterogeneous, and often only partially paired across modalities. Foundation models offer a promising direction for learning transferable representations from such data, yet directly adapting foundation model ideas to single-cell biology remains challenging due to batch effects, modality mismatch, limited labels, tissue-specific context, and the distinctive structure of cellular systems. This dissertation develops computational foundations for single-cell foundation models through a progression from reproducible benchmarking to increasingly general and transferable modeling frameworks. A standardized benchmark and software platform is introduced for single-cell deep learning, enabling systematic evaluation across datasets, tasks, metrics, and model implementations. Transformer-based multimodal prediction models are then studied to learn interactions among cells, genes, and proteins while incorporating biological structure. Diverse single-cell analysis tasks are further formulated as posterior distribution estimation problems, leading to a conditional generative modeling framework that supports prediction, generation, transfer, and adaptation. Finally, cellular pretraining strategies are investigated to move beyond isolated cell profiles by incorporating inter- cell relationships, spatially resolved transcriptomic information, and probabilistic inductive bias. Together, these studies contribute to the development of reproducible, expressive, generalizable, and transferable models for single-cell biology. They show that progress toward single-cell founda- tion models requires not only larger datasets and model architectures, but also reliable evaluation, biologically meaningful structure, flexible learning objectives, and pretraining strategies tailored to the organization and heterogeneity of biological data.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf","1 online resource (140 p.) : ill."]},{"key":"dc:title","label":"Title","values":["Towards Foundation Model for Single-Cell Biology"]}]}],"canonical_facts":{"dc:contributor":["Xie, Yuying","Cui, Yuehua","Hu, Guanyu","Tan, Pang-Ning"],"dc:creator":["Tang, Wenzhuo"],"dc:date":["2026"],"dc:description":["Thesis (Ph.D.)--Michigan State University. Statistics - Doctor of Philosophy, 2026","Single-cell and spatial profiling technologies have created new opportunities to study biologicalsystems at cellular resolution, generating large-scale datasets across diverse tissues, organisms, modalities, developmental stages, disease conditions, and perturbations. These data provide rich views of cellular states and biological context, but they are also high-dimensional, sparse, noisy, heterogeneous, and often only partially paired across modalities. Foundation models offer a promising direction for learning transferable representations from such data, yet directly adapting foundation model ideas to single-cell biology remains challenging due to batch effects, modality mismatch, limited labels, tissue-specific context, and the distinctive structure of cellular systems. This dissertation develops computational foundations for single-cell foundation models through a progression from reproducible benchmarking to increasingly general and transferable modeling frameworks. A standardized benchmark and software platform is introduced for single-cell deep learning, enabling systematic evaluation across datasets, tasks, metrics, and model implementations. Transformer-based multimodal prediction models are then studied to learn interactions among cells, genes, and proteins while incorporating biological structure. Diverse single-cell analysis tasks are further formulated as posterior distribution estimation problems, leading to a conditional generative modeling framework that supports prediction, generation, transfer, and adaptation. Finally, cellular pretraining strategies are investigated to move beyond isolated cell profiles by incorporating inter- cell relationships, spatially resolved transcriptomic information, and probabilistic inductive bias. Together, these studies contribute to the development of reproducible, expressive, generalizable, and transferable models for single-cell biology. They show that progress toward single-cell founda- tion models requires not only larger datasets and model architectures, but also reliable evaluation, biologically meaningful structure, flexible learning objectives, and pretraining strategies tailored to the organization and heterogeneity of biological data.","Description based on online resource. Title from PDF t.p. (Michigan State University Fedora Repository, viewed ).","Includes bibliographical references."],"dc:format":["application/pdf","1 online resource (140 p.) : ill."],"dc:identifier":["etd:53208","local:Tang_grad.msu_0128D_22237","https://doi.org/doi:10.25335/7xbe-ts29"],"dc:language":["eng"],"dc:relation":["Electronic Theses & Dissertations"],"dc:rights":["Attribution 4.0 International"],"dc:subject":["Statistics"],"dc:title":["Towards Foundation Model for Single-Cell Biology"],"dc:type":["text","Electronic dissertations"]},"updated_at":"2026-11-18T18:59:44Z"}