A comparative study of supervised and unsupervised machine learning algorithms applied to human microbiome

Background

The human microbiome, consisting of diverse bacte-rial, fungal, protozoan and viral species, exerts a profound influence on various physiological processes and disease susceptibility. However, the complexity of microbiome data has presented significant challenges in the analysis and interpretation of these intricate datasets, leading to the development of specialized software that employs machine learning algorithms for these aims.

Methods

In this paper, we analyze raw data taken from 16S rRNA gene sequencing from three studies, including stool samples from healthy control, patients with adenoma, and patients with colorectal cancer. Firstly, we use network-based methods to reduce dimensions of the dataset and consider only the most important features. In addition, we employ supervised machine learning algorithms to make prediction.

Results

Results show that graph-based techniques reduces dimen-sion from 255 up to 78 features with modularity score 0.73 based on different centrality measures. On the other hand, projection methods (non-negative matrix factorization and principal component analysis) reduce dimensions to 7 features. Furthermore, we apply supervised machine learning algorithms on the most important features obtained from centrality measures and on the ones obtained from projection methods, founding that the evaluation metrics have approximately the same scores when applying the algorithms on the entire dataset, on 78 feature and on 7 features.

Conclusions

This study demonstrates the efficacy of graph-based and projection methods in the interpretation for 16S rRNA gene sequencing data. Supervised machine learning on refined features from both approaches yields comparable predictive performance, emphasizing specific microbial features-bacteroides, prevotella, fusobacterium, lysinibacillus, blautia, sphingomonas, and faecalibacterium-as key in predicting patient conditions from raw data.

Negative emotions disrupt intentional synchronization during group sensorimotor interaction

Emotions play a fundamental role in human interactions and trigger responses in physiological, psychological, and behavioral modalities. Interpersonal coordination often entails attunement between individuals in various modalities. Previous research has elucidated the mechanisms of interpersonal synchronization and the emotions aroused by joint action: cardiac activity aligns in disputing marital couples, spectators share enjoyment in observing live dance performances, and joint finger-tapping evokes positive emotions. However, little is known about the impact of emotions on intentional interpersonal synchronization. To address this problem, we conducted an experiment in 2022 asking 60 participants to engage in a three-way finger-tapping synchronization task. We systematically induced emotional states (positive, neutral, and negative) with social comparison feedback using success-failure manipulations. An analysis of behavior synchronization using the Kuramoto order parameter revealed that negative emotion induction significantly diminished time spent in synchrony compared to positive induction. Moreover, the results exposed incremental struggles in attaining higher levels of synchronization (Q2-Q3) after the induction of negative emotions. These outcomes further substantiate the necessity of integrating the indices of agents' emotions into interpersonal synchronization and coordination models. We discuss the implications of this work for research on interpersonal emotion in joint action and applied outcomes in emotion-aware technologies and interventions. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

Unravelling individual rhythmic abilities using machine learning

Humans can easily extract the rhythm of a complex sound, like music, and move to its regular beat, like in dance. These abilities are modulated by musical training and vary significantly in untrained individuals. The causes of this variability are multidimensional and typically hard to grasp in single tasks. To date we lack a comprehensive model capturing the rhythmic fingerprints of both musicians and non-musicians. Here we harnessed machine learning to extract a parsimonious model of rhythmic abilities, based on behavioral testing (with perceptual and motor tasks) of individuals with and without formal musical training (n = 79). We demonstrate that variability in rhythmic abilities and their link with formal and informal music experience can be successfully captured by profiles including a minimal set of behavioral measures. These findings highlight that machine learning techniques can be employed successfully to distill profiles of rhythmic abilities, and ultimately shed light on individual variability and its relationship with both formal musical training and informal musical experiences.

Inhalation boosts perceptual awareness and decision speed

The emergence of consciousness is one of biology's biggest mysteries. During the past two decades, a major effort has been made to identify the neural correlates of consciousness, but in comparison, little is known about the physiological mechanisms underlying first-person subjective experience. Attention is considered the gateway of information to consciousness. Recent work suggests that the breathing phase (i.e., inhalation vs. exhalation) modulates attention, in such a way that attention directed toward exteroceptive information would increase during inhalation. One key hypothesis emerging from this work is that inhalation would improve perceptual awareness and near-threshold decision-making. The present study directly tested this hypothesis. We recorded the breathing rhythms of 30 humans performing a near-threshold decision-making task, in which they had to decide whether a liminal Gabor was tilted to the right or the left (objective decision task) and then to rate their perceptual awareness of the Gabor orientation (subjective decision task). In line with our hypothesis, the data revealed that, relative to exhalation, inhalation improves perceptual awareness and speeds up objective decision-making, without impairing accuracy. Overall, the present study builds on timely questions regarding the physiological mechanisms underlying consciousness and shows that breathing shapes the emergence of subjective experience and decision-making.NEW & NOTEWORTHY Breathing is a ubiquitous biological rhythm in animal life. However, little is known about its effect on consciousness and decision-making. Here, we measured the respiratory rhythm of humans performing a near-threshold discrimination experiment. We show that inhalation, compared with exhalation, improves perceptual awareness and accelerates decision-making while leaving accuracy unaffected.

More sleep, more milk

Breastfeeding is the best feeding method for infants, but this task is particularly challenging for mothers. Sleep time and quality are undeniably reduced in the postpartum period. No study has demonstrated the relationship between slow-wave sleep and lactation. Here, we discuss a unique experimental case during which the mother self-reported her sleep with a SUUNTO 9 watch and quantified her milk volume, blind to sleep parameters. This case report highlights an interesting strong correlation between stage N3 (slow-wave) sleep duration and milk production. It also demonstrates that this production is linked positively to self-reported sleepiness in the morning and breast tension and negatively to the number of awakenings. These results emphasize the need for preserving sleep, especially N3 sleep, during breastfeeding. Splitting nighttime infant care between parents, preserving the mother's sleep as much as possible during the first part of the night, could help improve lactation.

CITATION

Aerts C, Janaqi S, Cochen de Cock V. More sleep, more milk. J Clin Sleep Med. 2023;19(8):1563-1565.

Impact of emotion-laden acoustic stimuli on group synchronisation performance

The ability to synchronise with other people is a core socio-motor competence acquired during human development. In this study we aimed to understand the impact of individual emotional arousal on joint action performance. We asked 15 mixed-gender groups (of 4 individuals each) to participate in a digital, four-way movement synchronisation task. Participants shared the same physical space, but could not see each other during the task. In each trial run, every participant was induced with an emotion-laden acoustic stimulus (pre-selected from the second version of International Affective Digitized Sounds). Our data demonstrated that the human ability to synchronise is overall robust to fluctuations in individual emotional arousal, but performance varies in quality and movement speed as a result of valence of emotional induction (both on the individual and group level). We found that three negative inductions per group per trial led to a drop in overall group synchronisation performance (measured as the median and standard deviation of Kuramoto's order parameter-an index measuring the strength of synchrony between oscillators, in this study, players) in the 15 sec post-induction. We report that negatively-valenced inductions led to slower oscillations, whilst positive induction afforded faster oscillations. On the individual level of synchronisation performance we found an effect of empathetic disposition (higher competence linked to better performance during the negative induction condition) and of participant's sex (males displayed better synchronisation performance with others). We believe this work is a blueprint for exploring the frontiers of inextricably bound worlds of emotion and joint action, be it physical or digital.

Positive emotions foster spontaneous synchronisation in a group movement improvisation task

Emotions are a natural vector for acting together with others and are witnessed in human behaviour, perception and body functions. For this reason, studies of human-to-human interaction, such as multi-person motor synchronisation, are a perfect setting to disentangle the linkage of emotion with socio-motor interaction. And yet, the majority of joint action studies aiming at understanding the impact of emotions on multi-person performance resort to enacted emotions, the ones that are emulated based on the previous experience of such emotions, and almost exclusively focus on dyadic interaction. In addition, tasks chosen to study emotion in joint action are frequently characterised by a reduced number of physical dimensions to gain experimental control and subsequent facilitation in data analysis. Therefore, it is not clear how naturalistically induced emotions diffuse in more ecological interactions with other people and how emotions affect the process of interpersonal synchronisation. Here, we show that positive and negative emotions differently alter spontaneous human synchronous behaviour during a multi-person improvisation task. The study involved 39 participants organised in triads who self-reported liking improvisational activities (e.g., dancing). The task involved producing improvisational movements with the right hand. Participants were emotionally induced by manipulated social feedback involving a personal ranking score. Three-dimensional spatio-temporal data and cardiac activity were extracted and transformed into oscillatory signals (phases) to compute behavioural and physiological synchrony. Our results demonstrate that individuals induced with positive emotions, as opposed to negative emotions or a neutral state, maintained behavioural synchrony with other group members for a longer period of time. These findings contribute to the emerging shift of neuroscience of emotion and affective sciences towards the environment  of social significance where emotions appear the most—in interaction with others. Our study showcases a method of quantification of synchrony in an improvisational and interactive task based on a well-established Kuramoto model.

Classifying Idiopathic Rapid Eye Movement Sleep Behavior Disorder, Controls, and Mild Parkinson's Disease Using Gait Parameters

BACKGROUND

Subtle gait changes associated with idiopathic rapid eye movement sleep behavior disorder (iRBD) could allow early detection of subjects with future synucleinopathies.

OBJECTIVE

The aim of this study was to create a multiclass model, using statistical learning from probability distribution of gait parameters, to distinguish between patients with iRBD, healthy control subjects (HCs), and patients with Parkinson's disease (PD).

METHODS

Gait parameters were collected in 21 participants with iRBD, 21 with PD, and 21 HCs, matched for age, sex, and education level. Lasso sparse linear regression explored gait features able to classify the three groups.

RESULTS

The final model classified iRBD from HCs and from patients with PD equally well, with 95% accuracy, 100% sensitivity, and 90% specificity.

CONCLUSIONS

Gait parameters and a pretrained statistical model can robustly distinguish participants with iRBD from HCs and patients with PD. This could be used to screen subjects with future synucleinopathies in the general population and to identify a conversion threshold to PD. © 2022 International Parkinson and Movement Disorder Society.

Bridging the gap between emotion and joint action

Our daily human life is filled with a myriad of joint action moments, be it children playing, adults working together (i.e., team sports), or strangers navigating through a crowd. Joint action brings individuals (and embodiment of their emotions) together, in space and in time. Yet little is known about how individual emotions propagate through embodied presence in a group, and how joint action changes individual emotion. In fact, the multi-agent component is largely missing from neuroscience-based approaches to emotion, and reversely joint action research has not found a way yet to include emotion as one of the key parameters to model socio-motor interaction. In this review, we first identify the gap and then stockpile evidence showing strong entanglement between emotion and acting together from various branches of sciences. We propose an integrative approach to bridge the gap, highlight five research avenues to do so in behavioral neuroscience and digital sciences, and address some of the key challenges in the area faced by modern societies.

Decoding identity from motion: how motor similarities colour our perception of self and others

For more than 4 decades, it has been shown that humans are particularly sensitive to biological motion and extract socially relevant information from it such as gender, intentions, emotions or a person’s identity. A growing number of findings, however, indicate that identity perception is not always highly accurate, especially due to large inter-individual differences and a fuzzy self-recognition advantage compared to the recognition of others. Here, we investigated the self-other identification performance and sought to relate this performance to the metric properties of perceptual/physical representations of individual motor signatures. We show that identity perception ability varies substantially across individuals and is associated to the perceptual/physical motor similarities between self and other stimuli. Specifically, we found that the perceptual representations of postural signatures are veridical in the sense that closely reflects the physical postural trajectories and those similarities between people’ actions elicit numerous misattributions. While, on average, people can well recognize their self-generated actions, they more frequently attribute to themselves the actions of those acting in a similar way. These findings are consistent with the common coding theory and support that perception and action are tightly linked and may modulate each other by virtue of similarity.

Concurrent Changes of Brain Functional Connectivity and Motor Variability When Adapting to Task Constraints

In behavioral neuroscience, the adaptability of humans facing different constraints has been addressed on one side at the brain level, where a variety of functional networks dynamically support the same performance, and on the other side at the behavioral level, where fractal properties in sensorimotor variables have been considered as a hallmark of adaptability. To bridge the gap between the two levels of observation, we have jointly investigated the changes of network connectivity in the sensorimotor cortex assessed by modularity analysis and the properties of motor variability assessed by multifractal analysis during a prolonged tapping task. Four groups of participants had to produce the same tapping performance while being deprived from 0, 1, 2, or 3 sensory feedbacks simultaneously (auditory and/or visual and/or tactile). Whereas tapping performance was not statistically different across groups, the number of brain networks involved and the degree of multifractality of the inter-tap interval series were significantly correlated, increasing as a function of feedback deprivation. Our findings provide first evidence that concomitant changes in brain modularity and multifractal properties characterize adaptations underlying unchanged performance. We discuss implications of our findings with respect to the degeneracy properties of complex systems, and the entanglement of adaptability and effective adaptation.

A framework for unifying ontology-based semantic similarity measures: a study in the biomedical domain

Ontologies are widely adopted in the biomedical domain to characterize various resources (e.g. diseases, drugs, scientific publications) with non-ambiguous meanings. By exploiting the structured knowledge that ontologies provide, a plethora of ad hoc and domain-specific semantic similarity measures have been defined over the last years. Nevertheless, some critical questions remain: which measure should be defined/chosen for a concrete application? Are some of the, a priori different, measures indeed equivalent? In order to bring some light to these questions, we perform an in-depth analysis of existing ontology-based measures to identify the core elements of semantic similarity assessment. As a result, this paper presents a unifying framework that aims to improve the understanding of semantic measures, to highlight their equivalences and to propose bridges between their theoretical bases. By demonstrating that groups of measures are just particular instantiations of parameterized functions, we unify a large number of state-of-the-art semantic similarity measures through common expressions. The application of the proposed framework and its practical usefulness is underlined by an empirical analysis of hundreds of semantic measures in a biomedical context.

The semantic measures library and toolkit: fast computation of semantic similarity and relatedness using biomedical ontologies

The semantic measures library and toolkit are robust open-source and easy to use software solutions dedicated to semantic measures. They can be used for large-scale computations and analyses of semantic similarities between terms/concepts defined in terminologies and ontologies. The comparison of entities (e.g. genes) annotated by concepts is also supported. A large collection of measures is available. Not limited to a specific application context, the library and the toolkit can be used with various controlled vocabularies and ontology specifications (e.g. Open Biomedical Ontology, Resource Description Framework). The project targets both designers and practitioners of semantic measures providing a JAVA library, as well as a command-line tool that can be used on personal computers or computer clusters.

AVAILABILITY AND IMPLEMENTATION

Downloads, documentation, tutorials, evaluation and support are available at http://www.semantic-measures-library.org.