FK506-binding protein-51 (FKBP51) is a molecular cochaperone recently shown to be a positive regulator of peroxisome proliferator-activated receptor (PPAR)γ, the master regulator of adipocyte differentiation and function. In cellular models of adipogenesis, loss of FKBP51 not only reduced PPARγ activity but also reduced lipid accumulation, suggesting that FKBP51 knock-out (KO) mice might have insufficient development of adipose tissue and lipid storage ability. This model was tested by examining wild-type (WT) and FKBP51-KO mice under regular and high-fat diet conditions. Under both diets, FKBP51-KO mice were resistant to weight gain, hepatic steatosis, and had greatly reduced white adipose tissue (WAT) but higher amounts of brown adipose tissue. Under high-fat diet, KO mice were highly resistant to adiposity and exhibited reduced plasma lipids and elevated glucose and insulin tolerance. Profiling of perigonadal and sc WAT revealed elevated expression of brown adipose tissue lineage genes in KO mice that correlated increased energy expenditure and a shift of substrate oxidation to carbohydrates, as measured by indirect calorimetry. To directly test PPARγ involvement, WT and KO mice were fed rosiglitazone agonist. In WT mice, rosiglitazone induced whole-body weight gain, increased WAT mass, a shift of substrate oxidation to lipids, and elevated expression of PPARγ-regulated lipogenic genes in WAT. In contrast, KO mice had reduced rosiglitazone responses for these parameters. Our results identify FKBP51 as an important regulator of PPARγ in WAT and as a potential new target in the treatment of obesity and diabetes.

Graphical abstract

We study the problem of semantic matching in product search, that is, given a customer query, retrieve all semantically related products from the catalog. Pure lexical matching via an inverted index falls short in this respect due to several factors: a) lack of understanding of hypernyms, synonyms, and antonyms, b) fragility to morphological variants (e.g. "woman" vs. "women"), and c) sensitivity to spelling errors. To address these issues, we train a deep learning model for semantic matching using customer behavior data. Much of the recent work on large-scale semantic search using deep learning focuses on ranking for web search. In contrast, semantic matching for product search presents several novel challenges, which we elucidate in this paper. We address these challenges by a) developing a new loss function that has an inbuilt threshold to differentiate between random negative examples, impressed but not purchased examples, and positive examples (purchased items), b) using average pooling in conjunction with n-grams to capture short-range linguistic patterns, c) using hashing to handle out of vocabulary tokens, and d) using a model parallel training architecture to scale across 8 GPUs. We present compelling offline results that demonstrate at least 4.7% improvement in Recall@100 and 14.5% improvement in mean average precision (MAP) over baseline stateof-the-art semantic search methods using the same tokenization method. Moreover, we present results and discuss learnings from online A/B tests which demonstrate the efficacy of our method.

Summary Background Pioneering effort has been made to facilitate the recognition of pathology in malignancies based on whole‐slide images (WSIs) through deep learning approaches. It remains unclear whether we can accurately detect and locate basal cell carcinoma (BCC) using smartphone‐captured images. Objectives To develop deep neural network frameworks for accurate BCC recognition and segmentation based on smartphone‐captured microscopic ocular images (MOIs). Methods We collected a total of 8046 MOIs, 6610 of which had binary classification labels and the other 1436 had pixelwise annotations. Meanwhile, 128 WSIs were collected for comparison. Two deep learning frameworks were created. The ‘cascade’ framework had a classification model for identifying hard cases (images with low prediction confidence) and a segmentation model for further in‐depth analysis of the hard cases. The ‘segmentation’ framework directly segmented and classified all images. Sensitivity, specificity and area under the curve (AUC) were used to evaluate the overall performance of BCC recognition. Results The MOI‐ and WSI‐based models achieved comparable AUCs around 0·95. The ‘cascade’ framework achieved 0·93 sensitivity and 0·91 specificity. The ‘segmentation’ framework was more accurate but required more computational resources, achieving 0·97 sensitivity, 0·94 specificity and 0·987 AUC. The runtime of the ‘segmentation’ framework was 15·3 ± 3·9 s per image, whereas the ‘cascade’ framework took 4·1 ± 1·4 s. Additionally, the ‘segmentation’ framework achieved 0·863 mean intersection over union. Conclusions Based on the accessible MOIs via smartphone photography, we developed two deep learning frameworks for recognizing BCC pathology with high sensitivity and specificity. This work opens a new avenue for automatic BCC diagnosis in different clinical scenarios. What's already known about this topic? The diagnosis of basal cell carcinoma (BCC) is labour intensive due to the large number of images to be examined, especially when consecutive slide reading is needed in Mohs surgery. Deep learning approaches have demonstrated promising results on pathological image‐related diagnostic tasks. Previous studies have focused on whole‐slide images (WSIs) and leveraged classification on image patches for detecting and localizing breast cancer metastases. What does this study add? Instead of WSIs, microscopic ocular images (MOIs) photographed from microscope eyepieces using smartphone cameras were used to develop neural network models for recognizing BCC automatically. The MOI‐ and WSI‐based models achieved comparable areas under the curve around 0·95. Two deep learning frameworks for recognizing BCC pathology were developed with high sensitivity and specificity. Recognizing BCC through a smartphone could be considered a future clinical choice.

Nemo-like kinase (NLK), an evolutionarily conserved serine/threonine kinase, is highly expressed in the brain, but its function in the adult brain remains not well understood. In this study, we identify NLK as an interactor of huntingtin protein (HTT). We report that NLK levels are significantly decreased in HD human brain and HD models. Importantly, overexpression of NLK in the striatum attenuates brain atrophy, preserves striatal DARPP32 levels and reduces mutant HTT (mHTT) aggregation in HD mice. In contrast, genetic reduction of NLK exacerbates brain atrophy and loss of DARPP32 in HD mice. Moreover, we demonstrate that NLK lowers mHTT levels in a kinase activity-dependent manner, while having no significant effect on normal HTT protein levels in mouse striatal cells, human cells and HD mouse models. The NLK-mediated lowering of mHTT is associated with enhanced phosphorylation of mHTT. Phosphorylation defective mutation of serine at amino acid 120 (S120) abolishes the mHTT-lowering effect of NLK, suggesting that S120 phosphorylation is an important step in the NLK-mediated lowering of mHTT. A further mechanistic study suggests that NLK promotes mHTT ubiquitination and degradation via the proteasome pathway. Taken together, our results indicate a protective role of NLK in HD and reveal a new molecular target to reduce mHTT levels.