Upcycling of nutrients from kitchen waste: Integration of anaerobic digestion system and microbial protein production system.

To upcycle the nutrients from kitchen waste (KW), an integrated system consisting of anaerobic digestion (AD) reactor and microbial protein (MP) production reactor was established in this study. The subsystem I (AD system) demonstrated an efficient bio-energy production (282.37 mL CH4/g VS), with 553.54 mg/L of NH4+-N remained in the digestate. The subsystem II (MP production system) utilized the nitrogenous constituents of the digestate, with 2.04 g/L MP production. In order to further enhance the recovery efficiency, C/N ratio in the subsystem II was studied. NH4+-N recovery efficiency was 23.08% higher after C/N ratio optimization along with 0.24 g/L increment on MP production. Over 0.7 g/L of essential amino acids was obtained, according with the qualitative necessary for the feeds. Also, the key enzyme abundance of CO2 releasing and amino acid biosynthesis was obviously increased with max. 55.21%. Meanwhile, the integrated system was profitable via a simplified economic assessment.

In-situ microbial protein production by using nitrogen extracted from multifunctional bio-electrochemical system.

Upgrading of waste nitrogen sources is considered as an important approach to promote sustainable development. In this study, a multifunctional bio-electrochemical system with three chambers was established, innovatively achieving 2.02 g/L in-situ microbial protein (MP) production via hydrogen-oxidizing bacteria (HOB) in the protein chamber (middle chamber), along with over 2.9 L CO2/(L·d) consumption rate. Also, 69% chemical oxygen demand was degraded by electrogenic bacteria in the anode chamber, resulting in the 394.67 J/L electricity generation. Focusing on the NH4+-N migration in the system, the current intensity contributed 4%-9% in the anode and protein chamber, whereas, the negative effect of -6.69% on contribution was shown in the cathode chamber. On the view of kinetics, NH4+-N migration in anode and cathode chambers was fitted well with Levenberg-Marquardt equation (R2 > 0.92), along with the well-matched results of HOB growth in the protein chamber based on Gompertz model (R2 > 0.99). Further evaluating MPs produced by HOB, 0.45 g/L essential amino acids was detected, showing the better amino acid profile than fish and soybean. Multifunctional bio-electrochemical system revealed the economic potential of producing 6.69 €/m3 wastewater according to a simplified economic evaluation.

HSF1 activates the FOXO3a-ΔNp63α-CDK4 axis to promote head and neck squamous cell carcinoma cell proliferation and tumour growth.

Head and neck squamous cell carcinoma (HNSCC) is one of the most prevalent cancers worldwide. Heat shock factor 1 (HSF1) is a conserved transcriptional factor that plays a critical role in maintaining cellular proteostasis. However, the role of HSF1 in HNSCC development remains largely unclear. Here, we report that HSF1 promotes forkhead box protein O3a (FOXO3a)-dependent transcription of ΔNp63α (p63 isoform in the p53 family; inhibits cell migration, invasion, and metastasis), which leads to upregulation of cyclin-dependent kinase 4 expression and HNSCC tumour growth. Ablation of HSF1 or treatment with KRIBB11, a specific pharmacological inhibitor of HSF1, significantly suppresses ΔNp63α expression and HNSCC tumour growth. Clinically, the expression of HSF1 is positively correlated with the expression of ΔNp63α in HNSCC tumours. Together, this study demonstrates that the HSF1-ΔNp63α pathway is critically important for HNSCC tumour growth.

Learning Fast and Slow: Propedeutica for Real-Time Malware Detection.

Existing malware detectors on safety-critical devices have difficulties in runtime detection due to the performance overhead. In this article, we introduce Propedeutica, a framework for efficient and effective real-time malware detection, leveraging the best of conventional machine learning (ML) and deep learning (DL) techniques. In Propedeutica, all software start executions are considered as benign and monitored by a conventional ML classifier for fast detection. If the software receives a borderline classification from the ML detector (e.g., the software is 50% likely to be benign and 50% likely to be malicious), the software will be transferred to a more accurate, yet performance demanding DL detector. To address spatial-temporal dynamics and software execution heterogeneity, we introduce a novel DL architecture (DeepMalware) for Propedeutica with multistream inputs. We evaluated Propedeutica with 9115 malware samples and 1338 benign software from various categories for the Windows OS. With a borderline interval of [30%, 70%], Propedeutica achieves an accuracy of 94.34% and a false-positive rate of 8.75%, with 41.45% of the samples moved for DeepMalwareanalysis. Even using only CPU, Propedeutica can detect malware within less than 0.1 s.

Adversarial Examples: Attacks and Defenses for Deep Learning.

With rapid progress and significant successes in a wide spectrum of applications, deep learning is being applied in many safety-critical environments. However, deep neural networks (DNNs) have been recently found vulnerable to well-designed input samples called adversarial examples. Adversarial perturbations are imperceptible to human but can easily fool DNNs in the testing/deploying stage. The vulnerability to adversarial examples becomes one of the major risks for applying DNNs in safety-critical environments. Therefore, attacks and defenses on adversarial examples draw great attention. In this paper, we review recent findings on adversarial examples for DNNs, summarize the methods for generating adversarial examples, and propose a taxonomy of these methods. Under the taxonomy, applications for adversarial examples are investigated. We further elaborate on countermeasures for adversarial examples. In addition, three major challenges in adversarial examples and the potential solutions are discussed.

GRAM-CNN: a deep learning approach with local context for named entity recognition in biomedical text.

Motivation: Best performing named entity recognition (NER) methods for biomedical literature are based on hand-crafted features or task-specific rules, which are costly to produce and difficult to generalize to other corpora. End-to-end neural networks achieve state-of-the-art performance without hand-crafted features and task-specific knowledge in non-biomedical NER tasks. However, in the biomedical domain, using the same architecture does not yield competitive performance compared with conventional machine learning models. Results: We propose a novel end-to-end deep learning approach for biomedical NER tasks that leverages the local contexts based on n-gram character and word embeddings via Convolutional Neural Network (CNN). We call this approach GRAM-CNN. To automatically label a word, this method uses the local information around a word. Therefore, the GRAM-CNN method does not require any specific knowledge or feature engineering and can be theoretically applied to a wide range of existing NER problems. The GRAM-CNN approach was evaluated on three well-known biomedical datasets containing different BioNER entities. It obtained an F1-score of 87.26% on the Biocreative II dataset, 87.26% on the NCBI dataset and 72.57% on the JNLPBA dataset. Those results put GRAM-CNN in the lead of the biological NER methods. To the best of our knowledge, we are the first to apply CNN based structures to BioNER problems. Availability and implementation: The GRAM-CNN source code, datasets and pre-trained model are available online at: https://github.com/valdersoul/GRAM-CNN. Contact: andyli@ece.ufl.edu or aconesa@ufl.edu. Supplementary information: Supplementary data are available at Bioinformatics online.

Photonic crystal waveguide-based Mach-Zehnder demultiplexer.

A Mach-Zehnder demultiplexer with two different waveguide arms is proposed and studied theoretically in photonic crystal. The two waveguide arms with different widths function as a phase shifter. The operating wavelength spacing depends on the length of the two waveguide arms. The photonic bandgap is calculated by the plane-wave expansion method, and the device is simulated by the finite-difference time-domain method.