Search

Biomedical research often involves conducting experiments on model organisms in the anticipation that the biology learnt will transfer to humans. Previous comparative studies of mouse and human tissues were limited by the use of bulk-cell material. Here we show that transfer learning—the branch of machine learning that concerns passing information from one domain to another—can be used to efficiently map bone marrow biology between species, using data obtained from single-cell RNA sequencing. We first trained a multiclass logistic regression model to recognize different cell types in mouse bone marrow achieving equivalent performance to more complex artificial neural networks. Furthermore, it was able to identify individual human bone marrow cells with 83% overall accuracy. However, some human cell types were not easily identified, indicating important differences in biology. When re-training the mouse classifier using data from human, less than 10 human cells of a given type were needed to accurately learn its representation. In some cases, human cell identities could be inferred directly from the mouse classifier via zero-shot learning. These results show how simple machine learning models can be used to reconstruct complex biology from limited data, with broad implications for biomedical research. Patrick Stumpf et al. use a machine learning technique called transfer learning to compare bone marrow cell-type information between mice and humans, based on single-cell RNA-seq data. Using their model, they identify aspects of cellular expression profiles that transfer and those that don't, which can be used to understand when mouse models of human disease are appropriate. [ABSTRACT FROM AUTHOR]

Background: Obesity is associated with low-grade inflammation and impaired immune response. Caloric restriction (CR) has been shown to inhibit inflammatory response and enhance cell-mediated immune function. Curcumin, the bioactive phenolic component of turmeric spice, is proposed to have anti-obesity and anti-inflammation properties while piperine, another bioactive phenolic compound present in pepper spice, can enhance the bioavailability and efficacy of curcumin. This study sought to determine if curcumin could potentiate CR's beneficial effect on immune and inflammatory responses in obesity developed in mice by feeding high-fat diet (HFD). Methods: Mice were fed a HFD for 22 wk and then randomized into 5 groups: one group remained on HFD ad libitum and the remaining 4 groups were fed a 10% CR (reduced intake of HFD by 10% but maintaining the same levels of micronutrients) in the presence or absence of curcumin and/or piperine for 5 wk, after which CR was increased to 20% for an additional 33 wk. At the end of the study, mice were sacrificed, and spleen cells were isolated. Cells were stimulated with T cell mitogens, anti-CD3/CD28 antibodies, or lipopolysaccharide to determine T cell proliferation, cytokine production, and CD4+ T cell subpopulations. Results: Compared to HFD control group, all CR mice, regardless of the presence of curcumin and/or piperine, had lower body weight and fat mass, lower levels of blood glucose and insulin, and fewer total spleen cells but a higher percentage of CD4+ T cells. Additionally, they demonstrated lower production of pro-inflammatory cytokines IL-1β and TNF-α, a trend toward lower IL-6, and lower production of PGE2, a lipid molecule with pro-inflammatory and T cell-suppressive properties. Mice with CR alone had higher splenocyte proliferation and IL-2 production, but this effect of CR was diminished by spice supplementation. CR alone or in combination with spice supplementation had no effect on production of cytokines IL-4, IL-10, IFN-γ, and IL-17, or the proportion of different CD4+ T cell subsets. Conclusion: CR on an HFD favorably impacts both metabolic and immune/inflammatory profiles; however, the presence of curcumin and/or piperine does not amplify CR's beneficial effects. [ABSTRACT FROM AUTHOR]

A letter to the editor is presented in response to the article "Worldwide Demographic Crossroad Nears," by Hank Lindborg in the August 2006 issue.