Prediction of antimicrobial resistance based on whole-genome sequencing and machine learning.

MOTIVATION: Antimicrobial resistance (AMR) is one of the biggest global problems threatening human and animal health. Rapid and accurate AMR diagnostic methods are thus very urgently needed. However, traditional antimicrobial susceptibility testing (AST) is time-consuming, low throughput and viable only for cultivable bacteria. Machine learning methods may pave the way for automated AMR prediction based on genomic data of the bacteria. However, comparing different machine learning methods for the prediction of AMR based on different encodings and whole-genome sequencing data without previously known knowledge remains to be done. RESULTS: In this study, we evaluated logistic regression (LR), support vector machine (SVM), random forest (RF) and convolutional neural network (CNN) for the prediction of AMR for the antibiotics ciprofloxacin, cefotaxime, ceftazidime and gentamicin. We could demonstrate that these models can effectively predict AMR with label encoding, one-hot encoding and frequency matrix chaos game representation (FCGR encoding) on whole-genome sequencing data. We trained these models on a large AMR dataset and evaluated them on an independent public dataset. Generally, RFs and CNNs perform better than LR and SVM with AUCs up to 0.96. Furthermore, we were able to identify mutations that are associated with AMR for each antibiotic. AVAILABILITY AND IMPLEMENTATION: Source code in data preparation and model training are provided at GitHub website (https://github.com/YunxiaoRen/ML-iAMR). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

High-Throughput Preparation and High-Throughput Detection of Polymer-Dispersed Liquid Crystals Based on Ink-Jet Printing and Grayscale Value Analysis.

In this paper, based on high-throughput technology, polymer dispersed liquid crystals (PDLC) composed of pentaerythritol tetra (2-mercaptoacetic acid) (PETMP), trimethylolpropane triacrylate (TMPTA), and polyethylene glycol diacrylate (PEGD 600) were investigated in detail. A total of 125 PDLC samples with different ratios were quickly prepared using ink-jet printing. Based on the method of machine vision to identify the grayscale level of samples, as far as we know, it is the first time to realize high-throughput detection of the electro-optical performance of PDLC samples, which can quickly screen out the lowest saturation voltage of batch samples. Additionally, we compared the electro-optical test results of manual and high-throughput preparation PDLC samples and discovered that they had very similar electro-optical characteristics and morphologies. This demonstrated the viability of PDLC sample high-throughput preparation and detection, as well as promising application prospects, and significantly increased the efficiency of PDLC sample preparation and detection. The results of this study will contribute to the research and application of PDLC composites in the future.

Programmable Chromism and Photoluminescence of Spiropyran-Based Liquid Crystalline Polymer with Tunable Glass Transition Temperature.

Spiropyran-based materials (SPBMs) can give responses to the stimulations induced by the light, heat, force, or pH, which have been used as triggers for many smart materials. Here, a cross-linkable SPBM containing mesogenic-units is synthesized, which is pale-colored, non-photoluminescent and non-mesogenic at a spiro form, but dark-colored, photoluminescent, and mesogenic at a merocyanine form. Moreover, the dynamic interconversion behavior of the form in the different chemical environments are distinct. Liquid crystalline polymers (LCPs) containing the SPBMs cross-linked via visible light, own a photoswitchable glass transition temperature (Tg ) and retain the switchable property; however, the SPBMs cross-linked via UV light will be locked at the MC state, because the molecular movement was frozen at the room temperature lower than the given Tg of the LCP. Thus, programmable chromism and photoluminescence based on the tunable Tg can be endowed to the functional materials prepared from the SPBMs.

Programmable Dynamic Information Storage Composite Film with Highly Sensitive Thermochromism and Gradually Adjustable Fluorescence.

The development of an integrated material system capable of effectively organizing and combining multisource information, such as dynamic pigmentary, structural, and fluorescent colors, is significant and challenging. Achieving such programmable dynamic information storage can considerably enhance the diversity and security of information deliveries. Here, a polymer-stabilized cholesteric liquid crystal system with highly temperature-sensitive structural color and light-sensitive pigmentary and fluorescence colors is presented. The prepared cholesteric liquid crystals (clcs) can reversibly change their structural color from red to blue within variational 3 °C near room temperature, and exhibit a gradually adjustable fluorescence which can transform from blue to pink and finally to bright red. All this dynamic information is programmable and tailored, hundreds of thousands of (>540 000) pattern combinations can easily be achieved by optical writing with a "bagua" pattern photomask. Therefore, if the corresponding code combinations to the pattern are assigned particular meanings, encrypted transmission of information with very high security can be achieved by utilizing applicable information encoding tables and decryption rules.

Predicting hosts and cross-species transmission of Streptococcus agalactiae by interpretable machine learning.

BACKGROUND: Streptococcus agalactiae, commonly known as Group B Streptococcus (GBS), exhibits a broad host range, manifesting as both a beneficial commensal and an opportunistic pathogen across various species. In humans, it poses significant risks, causing neonatal sepsis and meningitis, along with severe infections in adults. Additionally, it impacts livestock by inducing mastitis in bovines and contributing to epidemic mortality in fish populations. Despite its wide host spectrum, the mechanisms enabling GBS to adapt to specific hosts remain inadequately elucidated. Therefore, the development of a rapid and accurate method differentiates GBS strains associated with particular animal hosts based on genome-wide information holds immense potential. Such a tool would not only bolster the identification and containment efforts during GBS outbreaks but also deepen our comprehension of the bacteria's host adaptations spanning humans, livestock, and other natural animal reservoirs. METHODS AND RESULTS: Here, we developed three machine learning models-random forest (RF), logistic regression (LR), and support vector machine (SVM) based on genome-wide mutation data. These models enabled precise prediction of the host origin of GBS, accurately distinguishing between human, bovine, fish, and pig hosts. Moreover, we conducted an interpretable machine learning using SHapley Additive exPlanations (SHAP) and variant annotation to uncover the most influential genomic features and associated genes for each host. Additionally, by meticulously examining misclassified samples, we gained valuable insights into the dynamics of host transmission and the potential for zoonotic infections. CONCLUSIONS: Our study underscores the effectiveness of random forest (RF) and logistic regression (LR) models based on mutation data for accurately predicting GBS host origins. Additionally, we identify the key features associated with each GBS host, thereby enhancing our understanding of the bacteria's host-specific adaptations.

Preparation of Highly Durable Reverse-Mode Polymer-Stabilized Liquid Crystal Films with Polymer Walls.

Reverse-mode polymer-stabilized liquid crystal (PSLC) films have wide applications in smart windows for cars as well as buildings and dimming glasses due to their low haze, low energy consumption, and better safety in case of emergency power off. However, PSLC films usually have poor stability of electro-optical properties due to their low polymer content (ca. 5 wt %), and it still remains a challenging task to improve the stability and processability by increasing the polymer content in PSLC as the driving voltage might dramatically increase. In this work, a reverse-mode PSLC film with polymer walls was prepared, which showed excellent stability of electro-optical properties even after 150 000 cycles. The film was prepared through polymerization with a photomask, in which the monomers concentrated on specific areas to form patterned polymer walls. In this way, the polymer content could be increased dramatically and the anchoring effect would not be too strong, thus avoiding a sharp increase in the driving voltage. As a result, the desired reverse-mode film with high stability, relatively low driving voltage, and high contrast ratio was obtained. The effects of monomer compositions, curing temperature, UV light intensity, and the pattern of the photomask on the microstructures, as well as electro-optical performances of the films were carefully studied. This work provides a new idea for the preparation of reverse-mode electrically switchable light-transmittance controllable films with excellent stability and good electro-optical performance, which would broaden their application in smart cars, building windows, and dimming glasses for light management and potential energy saving.

Multi-label classification for multi-drug resistance prediction of Escherichia coli.

Antimicrobial resistance (AMR) is a global health and development threat. In particular, multi-drug resistance (MDR) is increasingly common in pathogenic bacteria. It has become a serious problem to public health, as MDR can lead to the failure of treatment of patients. MDR is typically the result of mutations and the accumulation of multiple resistance genes within a single cell. Machine learning methods have a wide range of applications for AMR prediction. However, these approaches typically focus on single drug resistance prediction and do not incorporate information on accumulating antimicrobial resistance traits over time. Thus, identifying multi-drug resistance simultaneously and rapidly remains an open challenge. In our study, we could demonstrate that multi-label classification (MLC) methods can be used to model multi-drug resistance in pathogens. Importantly, we found the ensemble of classifier chains (ECC) model achieves accurate MDR prediction and outperforms other MLC methods. Thus, our study extends the available tools for MDR prediction and paves the way for improving diagnostics of infections in patients. Furthermore, the MLC methods we introduced here would contribute to reducing the threat of antimicrobial resistance and related deaths in the future by improving the speed and accuracy of the identification of pathogens and resistance.

Deep Transfer Learning Enables Robust Prediction of Antimicrobial Resistance for Novel Antibiotics.

Antimicrobial resistance (AMR) has become one of the serious global health problems, threatening the effective treatment of a growing number of infections. Machine learning and deep learning show great potential in rapid and accurate AMR predictions. However, a large number of samples for the training of these models is essential. In particular, for novel antibiotics, limited training samples and data imbalance hinder the models' generalization performance and overall accuracy. We propose a deep transfer learning model that can improve model performance for AMR prediction on small, imbalanced datasets. As our approach relies on transfer learning and secondary mutations, it is also applicable to novel antibiotics and emerging resistances in the future and enables quick diagnostics and personalized treatments.

Force-Induced Synergetic Pigmentary and Structural Color Change of Liquid Crystalline Elastomer with Nanoparticle-Enhanced Mechanosensitivity.

The ability of some animals to rapidly change their colors can greatly improve their chances of escaping predators or hunting prey. A classic example is cephalopods, which can rapidly shift through a wide range of colors. This ability is based on the synergetic effect of the change of pigmentary and structural colors exhibited by their own two categories of color-changing cells: supernatant chromatophores offer various pigmentary colors and lower iridophores or leucophores reflect the different structural colors by adjusting their periodicities. Here, a mechanochromic liquid crystalline elastomer with force-induced synergetic pigmentary and structural color change, whose mechanosensitivity is enhanced by the stress-concentration induced by the doped nanoparticle, is presented. The materials have a large color-changing gamut and high mechanochromic sensitivity, which exhibit great potential in the field of mechanical detectors, sensors, and anti-counterfeiting materials.

Regulatory association of long noncoding RNAs and chromatin accessibility facilitates erythroid differentiation.

Erythroid differentiation is a dynamic process regulated by multiple factors, whereas the interaction between long noncoding RNAs (lncRNAs) and chromatin accessibility and its influence on erythroid differentiation remains unclear. To elucidate this interaction, we used hematopoietic stem cells, multipotent progenitor cells, common myeloid progenitor cells, megakaryocyte-erythroid progenitor cells, and erythroblasts from human cord blood as an erythroid differentiation model to explore the coordinated regulatory functions of lncRNAs and chromatin accessibility by integrating RNA-seq and ATAC-seq data. We revealed that the integrated network of chromatin accessibility and lncRNAs exhibits stage-specific changes throughout the erythroid differentiation process and that the changes at the erythroblast stage of maturation are dramatic. We identified a subset of stage-specific lncRNAs and transcription factors (TFs) that associate with chromatin accessibility during erythroid differentiation, in which lncRNAs are key regulators of terminal erythroid differentiation via an lncRNA-TF-gene network. LncRNA PCED1B-AS1 was revealed to regulate terminal erythroid differentiation by coordinating GATA1 dynamically binding to the chromatin and interacting with the cytoskeleton network during erythroid differentiation. DANCR, another lncRNA that is highly expressed at the megakaryocyte-erythroid progenitor cell stage, was verified to promote erythroid differentiation by compromising megakaryocyte differentiation and coordinating with chromatin accessibility and TFs, such as RUNX1. Overall, our results identify the associated network of lncRNAs and chromatin accessibility in erythropoiesis and provide novel insights into erythroid differentiation and abundant resources for further study.

Long noncoding RNA PCED1B-AS1 promotes erythroid differentiation coordinating with GATA1 and chromatin remodeling.

Erythropoiesis is a complex and sophisticated multi-stage process regulated by a variety of factors, including the transcription factor GATA1 and non-coding RNA. GATA1 is regarded as an essential transcriptional regulator promoting transcription of erythroid-specific genes-such as long non-coding RNAs (lncRNA). Here, we comprehensively screened lncRNAs that were potentially regulated by GATA1 in erythroid cells. We identified a novel lncRNA-PCED1B-AS1-and verified its role in promoting erythroid differentiation of K562 erythroid cells. We also predicted a model in which PCED1B-AS1 participates in erythroid differentiation via dynamic chromatin remodeling involving GATA1. The relationship between lncRNA and chromatin in the process of erythroid differentiation remains to be revealed, and in our study we have carried out preliminary explorations.