HDMC

Motivation: With the development of single-cell RNA sequencing (scRNA-seq) techniques, increasingly more large-scale gene expression datasets become available. However, to analyze datasets produced by different experiments, batch effects among different datasets must be considered. Although several methods have been recently published to remove batch effects in scRNA-seq data, two problems remain to be challenging and not completely solved: (i) how to reduce the distribution differences of different batches more accurately; and (ii) how to align samples from different batches to recover the cell type clusters. Results: We proposed a novel deep-learning approach, which is a hierarchical distribution-matching framework assisted with contrastive learning to address these two problems. Firstly, we design a hierarchical framework for distribution matching based on a deep autoencoder. This framework employs an adversarial training strategy to match the global distribution of different batches. This provides an improved foundation to further match the local distributions with a maximum mean discrepancy-based loss. For local matching, we divide cells in each batch into clusters and develop a contrastive learning mechanism to simultaneously align similar cluster pairs and keep noisy pairs apart from each other. This allows to obtain clusters with all cells of the same type (true positives), and avoid clusters with cells of different type (false positives). We demonstrate the effectiveness of our method on both simulated and real datasets. Results show that our new method significantly outperforms the state-of-the-art methods and has the ability to prevent overcorrection. Availability and implementation: The python code to generate results and figures in this article is available at https://github.com/zhanglabNKU/HDMC, the data underlying this article is also available at this github repository. Supplementary information: Supplementary data are available at Bioinformatics online.

Background: The diet of low-income Guatemalan populations is mostly plant-based, deficient in some vitamins and minerals, and rich in antinutritional compounds that reduce the bioavailability of several micronutrients. Objective. To describe the manner in which low-income Guatemalan women in rural and urban settings would prepare dishes for main meals using a high-density multimicronutrient condiment (HDMC) and to compare the nutrient density and contribution to the recommended dietary allowance (RDA) of the dishes with and without added HDMC. Methods: A sample of 30 women, 15 each from rural and urban households, were enrolled. The women were given 20 g of the HDMC and asked to prepare a dish at home, serve it to their families at a time of their own choosing, record the recipe and the amount of condiment added, and report these facts to an interviewer on the following day. The nutrient content of each dish was calculated from food-composition table values for raw, whole ingredients and the package label values for the HDMC. Results: For all dishes combined, the HDMC provided on average 78% of the total vitamins. The proportion of the total vitamins provided by the HDMC varied greatly among different dishes. Typically a single serving of a dish without added HDMC provided less than half of the RDA (vitamin B12 and folate) for children and adult women. The midday meal is the most important meal of the day and should provide at least half of the RDA of all essential nutrients. With the HDMC added, the dishes on average provided 2 to 10 times the RDA for nutrients such as vitamins B6 and C and niacin in the reference children, and just satisfied the RDA intake for corresponding nutrients in adult women. Conclusions: The proportion of the RDAs of micronutrients provided by this novel, micronutrient-rich condiment varies over a wide range, depending on idiosyncrasies of recipe ingredients, amounts of condiment added, individuals served, and age- and physiology-dependent requirements. HDMCs could provide an efficient way to deliver essential micronutrients to vulnerable populations.

In this study, the compressive test and four-point flexural test were carried out to explore the water stability as well as mechanical properties of high ductility magnesium phosphate cement-based composites (HDMC). The effects of ambient curing age (7 d and 28 d), water immersion age (7 d, 28 d, and 56 d), water/binder ratio (W/B), and magnesium oxide/potassium dihydrogen phosphate ratio (M/P) on the mechanical properties (compressive strength, first-crack strength, ultimate flexural strength, ductility index, and toughness index) and water stability of the HDMC were examined. The results showed that the 28-day ambient curing could lead to higher retention rates of strength, ductility, and toughness than 7-day ambient curing, indicating better water stability; however, it did not result in significant improvement in the mechanical properties of the HDMC. As the water immersion age increased, the mechanical properties of the HDMC with 7-day ambient curing showed an obvious downward trend; the mechanical properties of the HDMC with 28-day ambient curing did not show an obvious decrease and even could be increased in many cases, especially when the water immersion age was 56 days; and the change of water stability was consistent with that of the mechanical properties. If all indexes and their corresponding retention rates were considered comprehensively, the W/B ratio of 0.16 and the M/P ratio of 5 seemed to be the optimum values for the HDMC. The scanning electron microscopy analysis confirmed that the water immersion had a large adverse effect on the HDMC and thus reduced their mechanical properties.