Yi Yao | Search bioRxiv

bioRxiv2016

A Novel Family of Genomics Islands Across Multiple Species of Streptococcus

The genus Streptococcus is one of the most genomically diverse and important human and agricultural pathogens. The acquisition of genomic islands (GIs) plays a central role in adaptation to new hosts in the genus pathogens. The research presented here employs a comparative genomics approach to define a novel family of GIs in the genus Streptococcus which also appears across strains of the same species. Specifically, we identified 9 Streptococcus genomes out of 67 sequenced genomes analyzed, and we termed these as 15bp Streptococcus genomic islands, or 15SGIs, including i) insertion adjacent to the 3 end of ribosome l7/l12 gene, ii) large inserts of horizontally acquired DNA, and iii) the presence of mobility genes (integrase) and replication initiators. We have identified a novel family of 15SGIs and seems to be important in species differentiation and adaptation to new hosts. It plays an important role during strain evolution in the genus Streptococcus.

Microbiology↗

bioRxiv2016

DeepCyTOF: Automated Cell Classification of Mass Cytometry Data by Deep Learning and Domain Adaptation

Mass cytometry or CyTOF is an emerging technology for high-dimensional multiparameter single cell analysis that overcomes many limitations of fluorescence-based flow cytometry. New methods for analyzing CyTOF data attempt to improve automation, scalability, performance, and interpretation of data generated in large studies. Assigning individual cells into discrete groups of cell types (gating) involves time-consuming sequential manual steps, untenable for larger studies. We introduce DeepCyTOF, a standardization approach for gating, based on deep learning techniques. DeepCyTOF requires labeled cells from only a single sample. It is based on domain adaptation principles and is a generalization of previous work that allows us to calibrate between a target distribution and a source distribution in an unsupervised manner. We show that Deep-CyTOF is highly concordant (98%) with cell classification obtained by individual manual gating of each sample when applied to a collection of 16 biological replicates of primary immune blood cells, even when measured accross several instruments. Further, DeepCyTOF achieves very high accuracy on the semi-automated gating challenge of the FlowCAP-I competition as well as two CyTOF datasets generated from primary immune blood cells: (i)14 subjects with a history of infection with West Nile virus (WNV), (ii) 34 healthy subjects of different ages. We conclude that deep learning in general, and DeepCyTOF specifically, offers a powerful computational approach for semi-automated gating of CyTOF and flow cytometry data.

Bioinformatics↗