Hybrid Clustering of Single-Cell Gene Expression and Spatial Information via Integrated NMF and K-Means.

Advances in single cell transcriptomics have allowed us to study the identity of single cells. This has led to the discovery of new cell types and high resolution tissue maps of them. Technologies that measure multiple modalities of such data add more detail, but they also complicate data integration. We offer an integrated analysis of the spatial location and gene expression profiles of cells to determine their identity. We propose scHybridNMF (single-cell Hybrid Nonnegative Matrix Factorization), which performs cell type identification by combining sparse nonnegative matrix factorization (sparse NMF) with k-means clustering to cluster high-dimensional gene expression and low-dimensional location data. We show that, under multiple scenarios, including the cases where there is a small number of genes profiled and the location data is noisy, scHybridNMF outperforms sparse NMF, k-means, and an existing method that uses a hidden Markov random field to encode cell location and gene expression data for cell type identification.

Risk factors associated with head louse infestation in Korea.

Head louse infestation (HLI) is one of the most frequently occurring parasitic diseases in children. This study was conducted to investigate the socioeconomic and personal factors influencing HLI in the Republic of Korea. A total of 2,210 questionnaires about various factors related to HLI were obtained from children in 17 primary schools throughout the country. The rate of HLI was significantly lower in children who lived together with mother or in a family where both parents worked. In addition, HLI was lower in children whose fathers or mothers were public officers or teachers. However, HLI was higher in children who had small families and washed their hair less often. Education levels of parents and the number of children in family were not significant. Improvement of socioeconomic factors and personal hygiene will be helpful for reducing HLI.

The use of an in vitro adenosine triphosphate-based chemotherapy response assay to predict chemotherapeutic response in breast cancer.

The adenosine triphosphate-based chemotherapy response assay (ATP-CRA) has the advantages of standardization, evaluability, reproducibility, and accuracy, and can be performed on relatively small numbers of tumor cells. A total of 43 patients were enrolled in the present study, and chemosensitivity tests were successfully performed in 40 (93.0%) of these patients. Twenty of the 40 received neoadjuvant chemotherapy or chemotherapy for metastatic breast cancer. The chemotherapy regimens used were doxorubicin plus docetaxel (n=9, 45.0%) or doxorubicin plus paclitaxel (n=11, 55.0%). Mean cell death rate, as determined by ATP-CRA, was lower in non-responders than in responders to therapy (P=0.012). Sensitivity, specificity, positive predictive value, negative predictive value, and diagnostic accuracy for ATP-CRA were 78.6%, 100%, 100%, 66.7%, and 85.0%, respectively. Diagnostic accuracy achieved by immunohistochemistry using estrogen receptor or progesterone receptor was lower than that achieved using ATP-CRA. Expression of p53, erb-B2, Ki67, Bcl-2, Bcl-xL, and annexin I was not significantly associated with response to chemotherapy. Our results show that ATP-CRA has high specificity and positive predictive value for predicting response to chemotherapy.