Browsing by Author "Chen, Siyi"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Chronic Infection Depletes Hematopoietic Stem Cells through Stress-Induced Terminal Differentiation
    (Cell Press, 2016) Matatall, Katie A.; Jeong, Mira; Chen, Siyi; Sun, Deqiang; Chen, Fengju; Mo, Qianxing; Kimmel, Marek; King, Katherine Y.
    Chronic infections affect a third of the world’s population and can cause bone marrow suppression, a severe condition that increases mortality from infection. To uncover the basis for infection-associated bone marrow suppression, we conducted repeated infection of WT mice with Mycobacterium avium. After 4–6 months, mice became pancytopenic. Their hematopoietic stem and progenitor cells (HSPCs) were severely depleted and displayed interferon gamma (IFN-γ) signaling-dependent defects in self-renewal. There was no evidence of increased HSPC mobilization or apoptosis. However, consistent with known effects of IFN-γ, transcriptome analysis pointed toward increased myeloid differentiation of HSPCs and revealed the transcription factor Batf2 as a potential mediator of IFN-γ-induced HSPC differentiation. Gain- and loss-of-function studies uncovered a role for Batf2 in myeloid differentiation in both murine and human systems. We thus demonstrate that chronic infection can deplete HSPCs and identify BATF2 as a mediator of infection-induced HSPC terminal differentiation.
  • Thumbnail Image
    Item
    Statistical modeling for species count data with heterogeneity
    (2020-12-03) Chen, Siyi; Kimmel, Marek; King, Katherine Y
    Hematopoietic stem cells (HSCs) can give rise to all different types of mature blood cells, thus estimating the number of clones is crucial in assessing the robustness of observations made in experiments. We first adopt a uniform-size model which assumes that every clone size is equal and identify its posterior mode as an estimate for clone size. We improve the model by applying a uniform prior since it is more biologically reasonably that all barcodes are evenly distributed in the blood system. We further propose a symmetric Dirichlet-multinomial sampling model for barcoding population size estimation in which we applied newly developed approximate Bayesian computation (ABC) approaches to realize the parameter estimation. For species sampling problem the aim is to estimate the unobserved proportion in the population. Geometric distribution is very importance for modeling species sampling data however the heterogeneity in data may lead to biased results in the estimation. Inspired by the Poisson mixture model approach we hence proposed a mixed geometric model in this chapter. In order to estimate the size of the target pop- ulation as well as taking species latent heterogeneity into consideration, we assume that the species count follows beta-geometric distribution, which can address data heterogeneity caused by over-dispersion and under-dispersion. The geometric distri- bution is a special case of the negative binomial distribution, hence we also propose a species population size estimator based on the beta-negative binomial distribution. The beta-negative binomial distribution a kind of mixed Poisson distribution that addresses heterogeneity which also includes the geometric distribution, the Poisson distribution, the negative binomial distribution and the beta-geometric distribution. We also address the species problem by proposing two novel over-dispersed distributions.