A Critical Analysis of the Limitation of Deep Learning based 3D Dental Mesh Segmentation Methods in Segmenting Partial Scans.

Tooth segmentation from intraoral scans is a crucial part of digital dentistry. Many Deep Learning based tooth segmentation algorithms have been developed for this task. In most of the cases, high accuracy has been achieved, although, most of the available tooth segmentation techniques make an implicit restrictive assumption of full jaw model and they report accuracy based on full jaw models. Medically, however, in certain cases, full jaw tooth scan is not required or may not be available. Given this practical issue, it is important to understand the robustness of currently available widely used Deep Learning based tooth segmentation techniques. For this purpose, we applied available segmentation techniques on partial intraoral scans and we discovered that the available deep Learning techniques under-perform drastically. The analysis and comparison presented in this work would help us in understanding the severity of the problem and allow us to develop robust tooth segmentation technique without strong assumption of full jaw model.Clinical relevance- Deep learning based tooth mesh segmentation algorithms have achieved high accuracy. In the clinical setting, robustness of deep learning based methods is of utmost importance. We discovered that the high performing tooth segmentation methods under-perform when segmenting partial intraoral scans. In our current work, we conduct extensive experiments to show the extent of this problem. We also discuss why adding partial scans to the training data of the tooth segmentation models is non-trivial. An in-depth understanding of this problem can help in developing robust tooth segmentation tenichniques.

Multilocus nuclear markers provide new insights into the origin and evolution of the blackbuck (Antilope cervicapra, Bovidae).

Evolutionary relationships between members of the Antilopina taxon have been much debated in recent years. The 'true antelope' clade is currently comprised of 4 genera viz., Gazella, Nanger, Eudorcas and the monotypic genus Antilope, that includes A. cervicapra. Most studies have focused on the mitochondrial genome or morphological data to study their relationships. However, signals from mitochondrial data can often be misleading when compared with nuclear markers, as has been shown in multiple taxonomic groups. In this study, we revisit the phylogenetic relationships among members of Antilopina, particularly the phylogenetic position of A. cervicapra, using 12 nuclear markers and compare it with the mitochondrial tree. Furthermore, we explore the implications of the results of this study on the taxonomy and biogeography of Indian antelopes. The nuclear phylogenetic trees built using multiple coalescent and concatenated methods all supported a paraphyletic genus Gazella. Antilope was nested within Gazella as opposed to being sister to it, which was suggested by previous studies and our results based on mitochondrial markers. Our fossil-calibrated larger bovid phylogeny, based on nuclear markers, suggested that the Antilope lineage diverged from its sister species more recently in the Pleistocene, rather than in late Miocene as per previous studies. Our biogeographic analyses suggest that the lineage leading to genus Antilope dispersed into India from the Saharo-Arabian realm around 2 mya, post the expansion of grasslands. We speculate that the adaptations of this savanna-grassland specialist did not allow them to extend their range beyond the Indian subcontinent. Whereas, the only other true antelope in India, G. bennetti, extended its range into India more recently, probably after the establishment of the Thar desert in northwest India.

Phosphoinositide binding profiles of the PX domains of Giardia lamblia.

The membrane trafficking machinery that functions at the endomembrane system of Giardia lamblia appears to be significantly different from that present in most model eukaryotes. This machinery is important for encystation as cyst wall material is trafficked to the cell surface via encystation-specific vesicles. Since proteins containing the phosphoinositide-binding PX domains are known regulators of vesicular trafficking, BLAST search was used to identify the PX domains of G. lamblia. Six putative PX domain-containing ORFs were identified. Some of the encoded PX domains contained non-canonical amino acid residues in the highly conserved ligand binding pocket. In vitro and in vivo binding studies indicate that these domains have the ability to bind to diverse phosphoinositides. Also, coincidence detection is likely to play a significant role in ligand binding in vivo since domains that bind to the same lipid in vitro, exhibit differences in subcellular localization. Analyses of the expression of these six genes in trophozoites, encysting trophozoites and cysts showed that while the expression of four of the genes were downregulated in cysts, the other two were upregulated. The variation in ligand preference of the individual PX domains and the differential expression of most of the PX-domain encoding genes indicate that these PX domain-containing proteins are likely to perform diverse cellular functions.