PIFiA: self-supervised approach for protein functional annotation from single-cell imaging data

Fluorescence microscopy data describe protein localization patterns at single-cell resolution and have the potential to reveal whole-proteome functional information with remarkable precision. Yet, extracting biologically meaningful representations from cell micrographs remains a major challenge. Existing approaches often fail to learn robust and noise-invariant features or rely on supervised labels for accurate annotations. We developed PIFiA (Protein Image-based Functional Annotation), a self-supervised approach for protein functional annotation from single-cell imaging data. We imaged the global yeast ORF-GFP collection and applied PIFiA to generate protein feature profiles from single-cell images of fluorescently tagged proteins. We show that PIFiA outperforms existing approaches for molecular representation learning and describe a range of downstream analysis tasks to explore the information content of the feature profiles. Specifically, we cluster extracted features into a hierarchy of functional organization, study cell population heterogeneity, and develop techniques to distinguish multi-localizing proteins and identify functional modules. Finally, we confirm new PIFiA predictions using a colocalization assay, suggesting previously unappreciated biological roles for several proteins. Paired with a fully interactive website ( https://thecellvision.org/pifia/ ), PIFiA is a resource for the quantitative analysis of protein organization within the cell.

PIFiA: Self-supervised Approach for Protein Functional Annotation from Single-Cell Imaging Data

Fluorescence microscopy data describe protein localization patterns at single-cell resolution and have the potential to reveal whole-proteome functional information with remarkable precision. Yet, extracting biologically meaningful representations from cell micrographs remains a major challenge. Existing approaches often fail to learn robust and noise-invariant features or rely on supervised labels for accurate annotations. We developed PIFiA, (Protein Image-based Functional Annotation), a self-supervised approach for protein functional annotation from single-cell imaging data. We imaged the global yeast ORF-GFP collection and applied PIFiA to generate protein feature profiles from single-cell images of fluorescently tagged proteins. We show that PIFiA outperforms existing approaches for molecular representation learning and describe a range of downstream analysis tasks to explore the information content of the feature profiles. Specifically, we cluster extracted features into a hierarchy of functional organization, study cell population heterogeneity, and develop techniques to distinguish multi-localizing proteins and identify functional modules. Finally, we confirm new PIFiA predictions using a colocalization assay, suggesting previously unappreciated biological roles for several proteins. Paired with a fully interactive website (https://thecellvision.org/pifia/), PIFiA is a resource for the quantitative analysis of protein organization within the cell.

Automated analysis of high-content microscopy data with deep learning

Existing computational pipelines for quantitative analysis of high‐content microscopy data rely on traditional machine learning approaches that fail to accurately classify more than a single dataset without substantial tuning and training, requiring extensive analysis. Here, we demonstrate that the application of deep learning to biological image data can overcome the pitfalls associated with conventional machine learning classifiers. Using a deep convolutional neural network (DeepLoc) to analyze yeast cell images, we show improved performance over traditional approaches in the automated classification of protein subcellular localization. We also demonstrate the ability of DeepLoc to classify highly divergent image sets, including images of pheromone‐arrested cells with abnormal cellular morphology, as well as images generated in different genetic backgrounds and in different laboratories. We offer an open‐source implementation that enables updating DeepLoc on new microscopy datasets. This study highlights deep learning as an important tool for the expedited analysis of high‐content microscopy data.

Identification of LRP16 as a negative regulator of insulin action and adipogenesis in 3T3-L1 adipocytes

Leukemia related protein 16 (LRP16) was first cloned from acute myeloid leukemia cells in our laboratory. In the present study, we sought to investigate the role of LRP16 in insulin action and sensitivity, using LRP16-depleted and -overexpressing 3T3-L1 cells. LRP16 silencing resulted in a reduction of the expression and secretion of tumor necrosis factor-alpha (TNF-α) and a concomitant increase in the expression of peroxisome proliferator-activated receptor-gamma (PPAR-γ). Moreover, LRP16 depletion promoted insulin-induced glucose uptake and adipocyte differentiation of 3T3-L1 cells. In contrast, LRP16 overexpression increased TNF-α secretion, suppressed glucose uptake, and attenuated 3T3-L1 cell differentiation. The phosphorylation levels of insulin receptor substrate 1 (IRS-1), phosphatidylinositide 3-kinase (PI3-K), and Akt were increased in LRP16-deficient 3T3-L1 cells, and conversely, diminished in LRP16-overexpressing 3T3-L1 cells, when compared to the corresponding control cells. Additionally, LRP16 overexpression raised the phosphorylation level of mammalian target of rapamycin (mTOR). The pretreatment with rapamycin, a specific inhibitor of mTOR, prevented the TNF-α elevation and PPAR-γ reduction and restored the phosphorylation of IRS-1, PI3-K, and Akt in LRP16-overexpressing cells. Our data collectively indicate that LRP16 acts as a negative regulator of insulin action and adipogenesis in 3T3-L1 adipocytes, which involves the activation of the mTOR signaling pathway.

[Research on transparent apinoid enemator]

Transparent apinoid enemator is made of polymethacrylate material. It is composed of external shell, big cover, small cover, liquor drain tube and suspension belt. Lateral surface of the shell has 100-1500 ml volume mark. Liquor drain tube is made of PVC, its inner diameter is 6 mm. The cover can reduce contamination and maintain liquor temperature. The transparent enemator made by us can overcome the shortcomings of non-transparent enamel enemator which has been used for many years.

[Effect of interleukin-1, interleukin-6 on the intercellular communication of rat thyroid FRTL-5 cells]

We observed the effects of thyrotropin (TSH), interleukin (IL)-1 beta and IL-6 on the intercellular communication of FRTL-5 cells with fluorenscence redistribution after photobleaching (FRAP) analysis. FRTL-5 cells were cultured and exposed to the different concentrations of TSH, IL-1 beta and IL-6 for 12 hours. The mean fluorescence recovery rate (MFRR, %/min) of the cells labelled with carboxyfluocein diacetate (CFDA) after photobleaching was measured with laser scanning cytometry. The MFRR (%/min) of the cells after exposure to TSH was 0.445 +/- 0.033 at the control group, 0.679 +/- 0.054 at the group of 0.1 U/L, 0.950 +/- 0.073 at the group of 1 U/L, and 0.799 +/- 0.082 at the group of 5 U/L, respectively (F = 11.44, P < 0.01). The indicated that TSH could enhance the intercellular communication of FRTL-5 cells. The MFRR after exposure to IL-1 beta was 0.564 +/- 0.032 at the control group, 0.485 +/- 0.042 at the group of 10(3) U/L, 0.445 +/- 0.043 at the group of 10(4) U/L and 0.405 +/- 0.029 at the group of 10(5) U/L, respectively (F = 3.58, P < 0.01). The suggested that IL-1 beta could inhibite the intercellular communication of FRTL-5 cells. IL-6 had no obvious effect on the intercellular communication of FRTL-5 cells.