Opening the Random Forest Black Box of the Metabolome by the Application of Surrogate Minimal Depth.
Ontology highlight
ABSTRACT:
SUBMITTER: Wenck S
PROVIDER: S-EPMC8781913 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:MOTIVATION:It has been shown that the machine learning approach random forest can be successfully applied to omics data, such as gene expression data, for classification or regression and to select variables that are important for prediction. However, the complex relationships between predictor variables, in particular between causal predictor variables, make the interpretation of currently applied variable selection techniques difficult. RESULTS:Here we propose a new variable selection approach called surrogate minimal depth (SMD) that incorporates surrogate variables into the concept of minimal depth (MD) variable importance. Applying SMD, we show that simulated correlation patterns can be reconstructed and that the increased consideration of variable relationships improves variable selection. When compared with existing state-of-the-art methods and MD, SMD has higher empirical power to identify causal variables while the resulting variable lists are equally stable. In conclusion, SMD is a promising approach to get more insight into the complex interplay of predictor variables and outcome in a high-dimensional data setting. AVAILABILITY AND IMPLEMENTATION:https://github.com/StephanSeifert/SurrogateMinimalDepth. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.
| S-EPMC6761946 | biostudies-literature
Project description:Background This study reports on the main criteria used by Canadian cardiac surgery residency program committees (RPCs) to select applicants and the perceptions of Canadian medical students interested in cardiac surgery. Methods A 50-question online survey was sent to all 12 Canadian cardiac surgery RPCs. A similar 52-question online survey targeted at Canadian medical students interested in applying to cardiac surgery residency programs was distributed. Data from both surveys were analyzed using descriptive statistics. Results A total of 62% of all cardiac surgery RPC members (66 of 106) participated, including committee members from all 12 programs (range: 1-12 members per program; 9%-100% response rate per program) and 67% of program directors (8 of 12). Forty-one Canadian medical students (22 pre-clerks [54%], 2 MD/PhD students [5%], and 17 clinical clerks [41%]) participated. Committee members considered the following criteria to be most important when selecting candidates: on-service clinical performance, the interview, quality of reference letters from cardiac surgeons, and completing a rotation at the target program’s institution. In contrast, the following criteria relating to the candidate were considered to be less important: wanting to practice in the city or province of training, having a connection to the program location, and personally knowing committee members. Medical students’ perceptions were concordant regarding what factors are the most important but they overestimated the influence of non-clinical factors and research productivity in increasing their competitiveness. Conclusion Canadian cardiac surgery residency programs seek applicants who demonstrate clinical excellence, as assessed by surgical rotations and reference letters from colleagues, and strong interview performance. Graphical abstract
| S-EPMC9700213 | biostudies-literature
Project description:Dogs and cats in Brazil serve as primary hosts for a considerable number of parasites, which may affect their health and wellbeing. These may include endoparasites (e.g., protozoa, cestodes, trematodes, and nematodes) and ectoparasites (i.e., fleas, lice, mites, and ticks). While some dog and cat parasites are highly host-specific (e.g., Aelurostrongylus abstrusus and Felicola subrostratus for cats, and Angiostrongylus vasorum and Trichodectes canis for dogs), others may easily switch to other hosts, including humans. In fact, several dog and cat parasites (e.g., Toxoplasma gondii, Dipylidium caninum, Ancylostoma caninum, Strongyloides stercoralis, and Toxocara canis) are important not only from a veterinary perspective but also from a medical standpoint. In addition, some of them (e.g., Lynxacarus radovskyi on cats and Rangelia vitalii in dogs) are little known to most veterinary practitioners working in Brazil. This article is a compendium on dog and cat parasites in Brazil and a call for a One Health approach towards a better management of some of these parasites, which may potentially affect humans. Practical aspects related to the diagnosis, treatment, and control of parasitic diseases of dogs and cats in Brazil are discussed.
| S-EPMC3914713 | biostudies-literature
Project description:The basal ganglia (BG) is a brain structure that has long been proposed to play an essential role in action selection, and theoretical models of spiking neurons have tried to explain how the BG solves this problem. A recently proposed functional and biologically inspired network model of the striatum (an important nucleus of the BG) is based on spike-timing-dependent eligibility (STDE) and captured important experimental features of this nucleus. The model can recognize complex input patterns and consistently choose rewarded actions to respond to such sensory inputs. However, model tuning is challenging due to two main reasons. The first is the expert knowledge required, resulting in tedious and potentially biased trial-and-error procedures. The second is the computational cost of assessing model configurations (approximately 1.78 h per evaluation). This study addresses the model tuning problem through numerical optimization. Considering the cost of assessing solutions, the selected methods stand out due to their low requirements for solution evaluations and compatibility with high-performance computing. They are the SurrogateOpt solver of Matlab and the RBFOpt library, both based on radial basis function approximations, and DIRECT-GL, an enhanced version of the widespread black-box optimizer DIRECT. Besides, a parallel random search serves as a baseline reference of the outcome of opting for sophisticated methods. SurrogateOpt turns out to be the best option for tuning this kind of model. It outperforms, on average, the quality of the configuration found by an expert and works significantly faster and autonomously. RBFOpt and the random search share the second position, but their average results are below the option found by hand. Finally, DIRECT-GL follows this line becoming the worst-performing method.
| S-EPMC9630480 | biostudies-literature
Project description:In electronic health (eHealth) research, limited insight has been obtained on process outcomes or how the use of technology has contributed to the users' ability to have a healthier life, improved well-being, or activate new attitudes in their daily tasks. As a result, eHealth is often perceived as a black box. To open this black box of eHealth, methodologies must extend beyond the classic effect evaluations. The analyses of log data (anonymous records of real-time actions performed by each user) can provide continuous and objective insights into the actual usage of the technology. However, the possibilities of log data in eHealth research have not been exploited to their fullest extent. The aim of this paper is to describe how log data can be used to improve the evaluation and understand the use of eHealth technology with a broader approach than only descriptive statistics. This paper serves as a starting point for using log data analysis in eHealth research. Here, we describe what log data is and provide an overview of research questions to evaluate the system, the context, the users of a technology, as well as the underpinning theoretical constructs. We also explain the requirements for log data, the starting points for the data preparation, and methods for data collection. Finally, we describe methods for data analysis and draw a conclusion regarding the importance of the results for both scientific and practical applications. The analysis of log data can be of great value for opening the black box of eHealth. A deliberate log data analysis can give new insights into how the usage of the technology contributes to found effects and can thereby help to improve the persuasiveness and effectiveness of eHealth technology and the underpinning behavioral models.
| S-EPMC5565791 | biostudies-literature
Project description:A broad research programme in Arabidopsis thaliana has provided estimates of selection on specific alleles in specific contexts, and identified geographic patterns of alleles in genes linked to timing of flowering. A closely related field has successfully captured many key axes of the evolution of timing of flowering in other monocarpic species through statistical and demographic modelling of large empirical databases. There has as yet been no synthesis between these two fields. Here we examine ways in which the two fields inform each other, and how this synergy will shape our knowledge of life-history evolution as a whole.
| S-EPMC2819306 | biostudies-literature
Project description: Not available
| S-EPMC1129058 | biostudies-literature
Project description:ObjectivesA Gateway to Medicine programme, developed in partnership between a further and higher education setting and implemented to increase the socioeconomic diversity of medicine, was examined to identify precisely what works within the programme and why.DesignThis study employed realist evaluation principles and was undertaken in three phases: document analysis and qualitative focus groups with widening access (WA) programme architects; focus groups and interviews with staff and students; generation of an idea of what works.SettingParticipants were recruited from a further/higher education setting and were either enrolled or involved in the delivery of a Gateway to Medicine programme.ParticipantsTwelve staff were interviewed either individually (n=3) or in one of three group interviews. Nine focus groups (ranging from 5 to 18 participants in each focus group) were carried out with Gateway students from three consecutive cohorts at 2-3 points in their Gateway programme year.ResultsData were generated to determine what 'works' in the Gateway programme. Turning a realist lens on the data identified six inter-relating mechanisms which helped students see medicine as attainable and achievable and prepared them for the transition to medical school. These were academic confidence (M1); developing professional identity (M2); financial support/security (M3); supportive relationships with staff (M4) and peers (M5); and establishing a sense of belonging as a university student (M6).ConclusionsBy unpacking the 'black box' of a Gateway programme through realist evaluation, we have shown that such programmes are not solely about providing knowledge and skills but are rather much more complex in respect to how they work. Further work is needed to further test the mechanisms identified in our study in other contexts for theory development and to identify predictors of effectiveness in terms of students' preparedness to transition.
| S-EPMC8280889 | biostudies-literature
Project description:PurposeLittle is known about how health-care professionals communicate with patients about consenting to genome sequencing. We therefore examined what topics health-care professionals covered and what questions patients asked during consent conversations.MethodsTwenty-one genome sequencing consent appointments were audio recorded and analyzed. Participants were 35 individuals being invited to participate in the 100,000 Genomes Project (14 participants with rare diseases, 21 relatives), and 10 health-care professionals ("consenters").ResultsTwo-thirds of participants' questions were substantive (e.g., genetics and inheritance); one-third administrative (e.g., filling in the consent form). Consenters usually (19/21) emphasized participant choice about secondary findings, but less often (13/21) emphasized the uncertainty about associated disease risks. Consenters primarily used passive statements and closed-ended, rather than open-ended, questions to invite participants' questions and concerns. In two appointments, one parent expressed negative or uncertain views about secondary findings, but after discussion with the other parent opted to receive them.ConclusionHealth-care professionals need to be prepared to answer patients' questions about genetics to facilitate genome sequencing consent. Health-care professionals' education also needs to address how to effectively listen and elicit each patient's questions and views, and how to discuss uncertainty around the disease risks associated with secondary findings.
| S-EPMC6752270 | biostudies-literature
Project description:(1) Background: tumor profiling enables patient survival prediction. The two essential parameters to be calibrated when designing a study based on tumor profiles from a cohort are the sequencing depth of RNA-seq technology and the number of patients. This calibration is carried out under cost constraints, and a compromise has to be found. In the context of survival data, the goal of this work is to benchmark the impact of the number of patients and of the sequencing depth of miRNA-seq and mRNA-seq on the predictive capabilities for both the Cox model with elastic net penalty and random survival forest. (2) Results: we first show that the Cox model and random survival forest provide comparable prediction capabilities, with significant differences for some cancers. Second, we demonstrate that miRNA and/or mRNA data improve prediction over clinical data alone. mRNA-seq data leads to slightly better prediction than miRNA-seq, with the notable exception of lung adenocarcinoma for which the tumor miRNA profile shows higher predictive power. Third, we demonstrate that the sequencing depth of RNA-seq data can be reduced for most of the investigated cancers without degrading the prediction abilities, allowing the creation of independent validation sets at a lower cost. Finally, we show that the number of patients in the training dataset can be reduced for the Cox model and random survival forest, allowing the use of different models on different patient subgroups.
| S-EPMC9777708 | biostudies-literature