Cellular heterogeneity and plasticity are central to development, regeneration, disease progression, and cell engineering. Advances in single-cell sequencing and large-scale cell cartography have revealed extensive transcriptional diversity, yet fully resolving tissue composition across physiological and pathological contexts remains challenging, particularly for underrepresented populations such as human bone marrow mesenchymal stem/stromal cells (MSCs). In parallel, substantial progress has been made in elucidating the molecular determinants and regulatory logic of gene transcription, including how transcription factors (TFs) engage chromatin and interpret regulatory DNA. However, our ability to quantitatively predict regulatory outcomes across the full spectrum of cell states remains limited. This gap is especially evident in TF-driven cell reprogramming, where the influence of TF dose on gene expression programs and cell fate transitions remains poorly understood. To address these challenges, this thesis develops experimental and analytical frameworks for dissecting TF dose-resolved regulatory activity at single-cell resolution. Central to this work is scTF-seq, a barcoded, doxycycline-inducible TF overexpression technique that enables quantitative mapping of transcriptomic responses across wide TF dose ranges. Applied to mouse embryonic MSCs, scTF-seq generated a gain-of-function atlas for 384 TFs, identifying previously uncharacterized TF functions, highlighting both monotonic and non-monotonic dose-response patterns that drive dose-dependent and/or stochastic reprogramming heterogeneity, and classifying TFs into distinct reprogramming capacity and dose sensitivity categories. Combinatorial scTF-seq experiments further revealed complex, dose-dependent TF interactions, including synergistic and antagonistic effects contingent on relative TF levels.

Building on these findings, we propose extending scTF-seq to multimodal, temporal, and spatial single-cell profiling. Such integration would directly link TF dose with chromatin accessibility, cis-regulatory element activity, dynamic gene regulatory network architecture, and the spatial and biophysical organization of transcriptional machinery, providing mechanistic insights into how TF dose shapes gene regulation and cellular heterogeneity. Finally, we apply single-cell transcriptomic profiling to human bone marrow MSCs enriched from aged controls and leukemia-remodeled marrows, generating a high-resolution MSC atlas. This complementary work reveals functionally polarized MSC states, disease-associated MSC remodeling, and culture-induced suppression of transcriptional diversity. Together, this work establishes a quantitative framework for decoding how TF dose and environmental context govern cellular heterogeneity and plasticity, enabling more precise and predictable engineering of cell states.