Search

Our lineage tracing studies using multiple Cre mouse lines showed a concurrent labeling of abundant taste bud cells and the underlying connective tissue with a neural crest (NC) origin, warranting a further examination on the issue of whether there is an NC derivation of taste bud cells. In this study, we mapped NC cell lineages in three different models, Sox10-iCreER T2 /tdT mouse, GFP+ neural fold transplantation to GFP− chickens, and Sox10-Cre/GFP-RFP zebrafish model. We found that in mice, Sox10-iCreER T2 specifically labels NC cell lineages with a single dose of tamoxifen at E7.5 and that the labeled cells were widely distributed in the connective tissue of the tongue. No labeled cells were found in taste buds or the surrounding epithelium in the postnatal mice. In the GFP+/GFP− chicken chimera model, GFP+ cells migrated extensively to the cranial region of chicken embryos ipsilateral to the surgery side but were absent in taste buds in the base of oral cavity and palate. In zebrafish, Sox10-Cre/GFP-RFP faithfully labeled known NC-derived tissues but did not label taste buds in lower jaw or the barbel. Our data, together with previous findings in axolotl, indicate that taste buds are not derived from NC cells in rodents, birds, amphibians or teleost fish. • Neural crest-derived cells are not seen in taste buds in mouse, chicken, and zebrafish. • Sox10-iCreER T2 is ideal for neural crest lineage mapping in mouse. • GFP+/GFP− chicken chimera is valuable for longitudinally tracking neural crest cells. [ABSTRACT FROM AUTHOR]

In 3D video extension of High-Efficiency-Video-Coding (3D-HEVC), intra depth coding involves huge computational complexity for the added depth modeling modes (DMMs) and the new complex processing flow. This paper proposes an algorithm including the fast mode and prediction unit (PU) size decisions to accelerate the intra depth coding. For the mode decision, after a modification of the processing flow, three strategies are adopted: (1) evaluation results for intra conventional modes (ICMs) are utilized to determine whether DMMs are skipped; (2) with a predicted direction from ICMs, a golden ratio method is adopted to simplify the DMMs searching; and (3) fast Planar mode decision is made by analyzing the relationship of DMMs and Planar mode. For the fast PU size decision, evaluation costs of ICMs and DMMs, with combination of some trained thresholds, are used to make decisions. Experimental results show the overall algorithm reduces 63.00% computational complexity, with 1.04% BD-rate increasing. [ABSTRACT FROM AUTHOR]