Artificial Neural Networks and Deep Learning for Genomic Prediction of Binary, Ordinal, and Mixed Outcomes

In this chapter, we provide the main elements for implementing deep neural networks in Keras for binary, categorical, and mixed outcomes under feedforward networks as well as the main practical issues involved in implementing deep learning models with binary response variables. The same practical issues are provided for implementing deep neural networks with categorical and count traits under a univariate framework. We follow with a detailed assessment of information for implementing multivariate deep learning models for continuous, binary, categorical, count, and mixed outcomes. In all the examples given, the data came from plant breeding experiments including genomic data. The training process for binary, ordinal, count, and multivariate outcomes is similar to fitting DNN models with univariate continuous outcomes, since once we have the data to be trained, we need to (a) define the DNN model in Keras, (b) configure and compile the model, (c) fit the model, and finally, (d) evaluate the prediction performance in the testing set. In the next section, we provide illustrative examples of training DNN for binary outcomes in Keras R (Chollet and Allaire, Deep learning with R. Manning Publications, Manning Early Access Program (MEA), 2017; Allaire and Chollet, Keras: R interface to Keras’, 2019).

Tocilizumab reduces COVID-19 mortality and pathology in a dose and timing-dependent fashion: a multi-centric study

Life-threatening COVID-19 is associated with strong inflammation, where an IL-6-driven cytokine storm appears to be a cornerstone for enhanced pathology. Nonetheless, the specific inhibition of such pathway has shown mixed outcomes. This could be due to variations in the dose of tocilizumab used, the stage in which the drug is administered or the severity of disease presentation. Thus, we performed a retrospective multicentric study in 140 patients with moderate to critical COVID-19, 79 of which received tocilizumab in variable standard doses (< 400 mg, 400–800 mg or > 800 mg), either at the viral (1–7 days post-symptom onset), early inflammatory (8–15) or late inflammatory (16 or more) stages, and compared it with standard treated patients. Mortality, reduced respiratory support requirements and pathology markers were measured. Tocilizumab significantly reduced the respiratory support requirements (OR 2.71, CI 1.37–4.85 at 95%) and inflammatory markers (OR 4.82, CI 1.4–15.8) of all patients, but mortality was only reduced (4.1% vs 25.7%, p = 0.03) when the drug was administered at the early inflammatory stage and in doses ranging 400–800 mg in severely-ill patients. Despite the apparent inability of Tocilizumab to prevent the progression of COVID-19 into a critical presentation, severely-ill patients may be benefited by its use in the early inflammatory stage and moderate doses.

Thematic Co-occurrence Analysis: Advancing a Theory and Qualitative Method to Illuminate Ambivalent Experiences

Ambivalence is a phenomenon that transcends disciplinary divides and is associated with a myriad of mixed outcomes. Yet, identifying and representing the complexities of ambivalent experiences can be difficult using traditional qualitative methods. Thus, the goal of the present study was to advance a qualitative method, thematic co-occurrence analysis, to address this issue. To illustrate the usefulness of this method, I present a case study detailing 35 estranged adult children’s ambivalent responses and reactions to their parents’ (non)contact during the COVID-19 pandemic. Findings reveal seven themes and four (non)relationships between them that reflect (a) theme independence, (b) unilateral co-occurrence, (c) unbalanced co-occurrence, and (d) complementary co-occurrence. These findings advance a theory of ambivalence and corresponding method to aid in the future investigations of ambivalent phenomena.

Talking about a Revolution

This chapter anatomizes the Mexican state’s attempt to foster a nationalist culture and to control a fragmented public sphere. It finds that Mexican politicians invested heavily in both to distinctly mixed outcomes. Governments at all levels spent on schools, rituals and propaganda materials that the ruled consumed, quite often enthusiastically, but with their eyes open and with judgement suspended as to how nationalist stories squared with reality. While acknowledging the power of content management in radio and television broadcasting it revises traditional appreciations of the print press. Mexican newspapers were neither docile nor wholly commercial, and provincial publications in particular could engage in systematic opposition to state, and on occasion national governments. Mexicans were selectively enthusiastic about parts of the state’s nationalist project, at its centre the revolution, but reliably sceptical as to how much of that project their politicians were actually delivering.

Contribution to the Special Issue on Clinical Ethology: Educating dog owners: how owner–dog interactions can benefit from addressing the human caregiving system and dog-directed parenting styles

Dogs need to adapt to a human environment to enhance their welfare and to avoid risks of undesired dog behaviour and relinquishment. Crucial to this adaptation may be how an owner interacts with the dog. Owner–dog interactions may be influenced by the human caregiving system with regard to how care, protection and resources are provided. This narrative review discusses how a consideration of the human caregiving system can benefit owner–dog interactions. Literature suggests that the human caregiving system and parenting styles could influence owner–dog interactions. Owner–dog education may improve these interactions. However, studies on owner–dog education present mixed outcomes for the dog. Also, only a few studies address owner outcomes, indicating a gap that needs filling. It is concluded that, when intervening in owner–dog interactions, more attention should be directed to aspects of human psychology. Dog-directed parenting styles can form one strategy as to improve owner–dog interactions and dog welfare.

Challenges and solutions for application and wider adoption of wearable robots

The science and technology of wearable robots are steadily advancing, and the use of such robots in our everyday life appears to be within reach. Nevertheless, widespread adoption of wearable robots should not be taken for granted, especially since many recent attempts to bring them to real-life applications resulted in mixed outcomes. The aim of this article is to address the current challenges that are limiting the application and wider adoption of wearable robots that are typically worn over the human body. We categorized the challenges into mechanical layout, actuation, sensing, body interface, control, human–robot interfacing and coadaptation, and benchmarking. For each category, we discuss specific challenges and the rationale for why solving them is important, followed by an overview of relevant recent works. We conclude with an opinion that summarizes possible solutions that could contribute to the wider adoption of wearable robots.

The Sibling-in-Law: Understanding an Unknown Member of the Disability Community

Little attention has focused on siblings-in-law, the spouses of siblings of people with disabilities. Using an online survey, 102 siblings-in-law provided information concerning themselves, their spouses, in-law family, and experiences. Overall experiences were rated as positive (51.0%), mixed (34.3%), or negative (14.7%). Compared to respondents with negative outcomes, respondents with positive and mixed outcomes felt being a sibling-in-law more positively impacted their marriage, relationship with in-laws, parenting, desire to advocate, and career. Before marriage, these two groups better understood sibling-in-law responsibilities and reported less worry about marrying into a family with a member with disabilities or having children with disabilities. In open-ended reflections, 1/3 of respondents discussed family dynamics; improving disability knowledge; future planning; and marriage considerations. Practice and research implications are discussed.

Regret in Experience-Based Decisions: The Effects of Expected Value Differences and Mixed Gains and Losses

Previous research on experience-based decisions with full feedback supports the idea that people tend to prefer options that minimize the probability of regret. The current study explored whether this preference is modulated by differences in expected value (EV) and the presence or absence of occasional losses. Participants (n = 52) completed an online experiment that involved repeated choices between a safer and a riskier option while receiving full feedback. The riskier option yielded a better outcome on 80% of draws so that choosing it minimized the probability of regret. Preference for the riskier, regret-minimizing option was high when it had the same EV as the safer option and all outcomes were gains, but it decreased when the safer option had a higher EV and when both options included occasional losses. Outcome ratings that were obtained on 50% of trials showed large effects of regret and rejoicing, confirming that participants were sensitive to relative comparisons between obtained and forgone outcomes. Reinforcement-learning modeling indicated that the effects of unequal EVs and mixed outcomes could be accounted for by assuming combined encoding of absolute and relative outcomes and unequal weighting of gains and losses. Overall, these results demonstrate that the impact of regret can be modulated by structural features of the choice environment.