Abstract
This survey paper discusses opportunities and threats of using artificial intelligence (AI) technology in the manufacturing sector with consideration for offensive and defensive uses of such technology. It starts with an introduction of Industry 4.0 concept and an understanding of AI use in this context. Then provides elements of security principles and detection techniques applied to operational technology (OT) which forms the main attack surface of manufacturing systems. As some intrusion detection systems (IDS) already involve some AI-based techniques, we focus on existing machine-learning and data-mining based techniques in use for intrusion detection. This article presents the major strengths and weaknesses of the main techniques in use. We also discuss an assessment of their relevance for application to OT, from the manufacturer point of view. Another part of the paper introduces the essential drivers and principles of Industry 4.0, providing insights on the advent of AI in manufacturing systems as well as an understanding of the new set of challenges it implies. AI-based techniques for production monitoring, optimisation and control are proposed with insights on several application cases. The related technical, operational and security challenges are discussed and an understanding of the impact of such transition on current security practices is then provided in more details. The final part of the report further develops a vision of security challenges for Industry 4.0. It addresses aspects of orchestration of distributed detection techniques, introduces an approach to adversarial/robust AI development and concludes with human-machine behaviour monitoring requirements.

Abstract
The significant growth of interconnected Internet-of-Things (IoT) devices, the use of social networks, along with the evolution of technology in different domains, lead to a rise in the volume of data generated continuously from multiple systems. Valuable information can be derived from these evolving data streams by applying machine learning. In practice, several critical issues emerge when extracting useful knowledge from these potentially infinite data, mainly because of their evolving nature and high arrival rate which implies an inability to store them entirely. In this work, we provide a comprehensive survey that discusses the research constraints and the current state-of-the-art in this vibrant framework. Moreover, we present an updated overview of the latest contributions proposed in different stream mining tasks, particularly classification, regression, clustering, and frequent patterns. This article is categorized under: Fundamental Concepts of Data and Knowledge > Key Design Issues in Data Mining Fundamental Concepts of Data and Knowledge > Motivation and Emergence of Data Mining

Abstract

Abstract
The continuous characterization of genome-wide diversity in population and case-cohort samples, allied to the development of new algorithms, are shedding light on host ancestry impact and selection events on various infectious diseases. Especially interesting are the long-standing associations between humans and certain bacteria, such as the case of Helicobacter pylori, which could have been strong drivers of adaptation leading to coevolution. Some evidence on admixed gastric cancer cohorts have been suggested as supporting Homo-Helicobacter coevolution, but reliable experimental data that control both the bacterium and the host ancestries are lacking. Here, we conducted the first in vitro coinfection assays with dual human- and bacterium-matched and -mismatched ancestries, in African and European backgrounds, to evaluate the genome wide gene expression host response to H. pylori. Our results showed that: (1) the host response to H. pylori infection was greatly shaped by the human ancestry, with variability on innate immune system and metabolism; (2) African human ancestry showed signs of coevolution with H. pylori while European ancestry appeared to be maladapted; and (3) mismatched ancestry did not seem to be an important differentiator of gene expression at the initial stages of infection as assayed here.

Abstract
Traditional detection of aquatic invasive species, via morphological identification is often time-consuming and can require a high level of taxonomic expertise, leading to delayed mitigation responses. Environmental DNA (eDNA) detection approaches of multiple species using Illumina-based sequencing technology have been used to overcome these hindrances, but sample processing is often lengthy. More recently, portable nanopore sequencing technology has become available, which has the potential to make molecular detection of invasive species more widely accessible and to substantially decrease sample turnaround times. However, nanopore-sequenced reads have a much higher error rate than those produced by Illumina platforms, which has so far hindered the adoption of this technology. We provide a detailed laboratory protocol and bioinformatic tools to increase the reliability of nanopore sequencing to detect invasive species, and we test its application using invasive bivalves. We sampled water from sites with pre-existing bivalve occurrence and abundance data, and contrasting bivalve communities, in Italy and Portugal. We extracted, amplified and sequenced eDNA with a turnaround of 3.5 days. The majority of processed reads were = 99 % identical to reference sequences. There were no taxa detected other than those known to occur. The lack of detections of some species at some sites could be explained by their known low abundances. The approach is now being tested on other target taxa such as fish and other vertebrates.

Abstract
Adjuvant systemic therapies are now routinely used following resection of stage III melanoma, however accurate prognostic information is needed to better stratify patients. We use differential expression analyses of primary tumours from 204 RNA-sequenced melanomas within a large adjuvant trial, identifying a 121 metastasis-associated gene signature. This signature strongly associated with progression-free (HR=1.63, p=5.24 x 10(-5)) and overall survival (HR=1.61, p=1.67 x 10(-4)), was validated in 175 regional lymph nodes metastasis as well as two externally ascertained datasets. The machine learning classification models trained using the signature genes performed significantly better in predicting metastases than models trained with clinical covariates (p(AUROC) = 7.03 x 10(-4)), or published prognostic signatures (p(AUROC) < 0.05). The signature score negatively correlated with measures of immune cell infiltration (=-0.75, p<2.2 x 10(-16)), with a higher score representing reduced lymphocyte infiltration and a higher 5-year risk of death in stage II melanoma. Our expression signature identifies melanoma patients at higher risk of metastases and warrants further evaluation in adjuvant clinical trials. The identification of prognostic biomarkers can help stratify cancer patients. Here, the authors apply deep RNA sequencing from primary melanomas coupled with long-term clinical outcome data from a prospective multicentre phase III trial, to develop and validate a 121 metastasis-associated gene signature identifying early-stage melanoma patients at higher risk of metastasis and worse survival.