Deep learning and support vector machines for transcription start site identification.

Recognizing transcription start sites is key to gene identification. Several approaches have been employed in related problems such as detecting translation initiation sites or promoters, many of the most recent ones based on machine learning. Deep learning methods have been proven to be exceptionally effective for this task, but their use in transcription start site identification has not yet been explored in depth. Also, the very few existing works do not compare their methods to support vector machines (SVMs), the most established technique in this area of study, nor provide the curated dataset used in the study. The reduced amount of published papers in this specific problem could be explained by this lack of datasets. Given that both support vector machines and deep neural networks have been applied in related problems with remarkable results, we compared their performance in transcription start site predictions, concluding that SVMs are computationally much slower, and deep learning methods, specially long short-term memory neural networks (LSTMs), are best suited to work with sequences than SVMs. For such a purpose, we used the reference human genome GRCh38. Additionally, we studied two different aspects related to data processing: the proper way to generate training examples and the imbalanced nature of the data. Furthermore, the generalization performance of the models studied was also tested using the mouse genome, where the LSTM neural network stood out from the rest of the algorithms. To sum up, this article provides an analysis of the best architecture choices in transcription start site identification, as well as a method to generate transcription start site datasets including negative instances on any species available in Ensembl. We found that deep learning methods are better suited than SVMs to solve this problem, being more efficient and better adapted to long sequences and large amounts of data. We also create a transcription start site (TSS) dataset large enough to be used in deep learning experiments.

Nonlinear physics opens a new paradigm for accurate transcription start site prediction.

There is evidence that DNA breathing (spontaneous opening of the DNA strands) plays a relevant role in the interactions of DNA with other molecules, and in particular in the transcription process. Therefore, having physical models that can predict these openings is of interest. However, this source of information has not been used before either in transcription start sites (TSSs) or promoter prediction. In this article, one such model is used as an additional information source that, when used by a machine learning (ML) model, improves the results of current methods for the prediction of TSSs. In addition, we provide evidence on the validity of the physical model, as it is able by itself to predict TSSs with high accuracy. This opens an exciting avenue of research at the intersection of statistical mechanics and ML, where ML models in bioinformatics can be improved using physical models of DNA as feature extractors.

Measuring the Functional Abilities of Children Aged 3-6 Years Old with Observational Methods and Computer Tools.

The analysis of functional abilities and their development in early childhood (0-6 years old) are fundamental aspects among young children with certain types of developmental difficulties that can facilitate prevention, through programmed interventions adapted to the needs of each user (student or patient). There are, however, few investigations to date, that have analyzed the use of automated tools for recording and interpreting the results of the initial assessment. Here, a protocol is presented to examine the functional abilities in early childhood in young children, aged between 3-6 years old, with intellectual disabilities, but the protocol can also be used for ages 0 to 6 years. The protocol makes use of a computer application, eEarlyCare, that facilitates the interpretation of the results of systematic observations, which are recorded in natural environments by professionals trained in early intervention. The software can be used to analyze 11 functional areas (Food Autonomy, Personal Care and Hygiene, Dressing and Undressing Independently, Sphincter Control, Functional Mobility, Communication and Language, Daily Life Routines, Adaptive Behavior and Attention) and a total of 114 different behaviors. Its use facilitates the analysis of the observed abilities and greatly assists early intervention. Compared to other observational methods, it allows a more efficient use of personal and material resources. The use of the computer application facilitates the recording of the observation results, which helps with organization and reflection on the observations. The software displays the observation results on-screen compared to normal developmental parameters. This information can be referred to for decision-making about the most suitable intervention program for each user (student or patient). Likewise, clustering techniques are applied to analyze the relation between the type of intellectual disabilities and functional development identified with the software, a relation that is intended to serve as a guide for early-care professional intervention.

Early and extremely early multi-label fault diagnosis in induction motors.

The detection of faulty machinery and its automated diagnosis is an industrial priority because efficient fault diagnosis implies efficient management of the maintenance times, reduction of energy consumption, reduction in overall costs and, most importantly, the availability of the machinery is ensured. Thus, this paper presents a new intelligent multi-fault diagnosis method based on multiple sensor information for assessing the occurrence of single, combined, and simultaneous faulty conditions in an induction motor. The contribution and novelty of the proposed method include the consideration of different physical magnitudes such as vibrations, stator currents, voltages, and rotational speed as a meaningful source of information of the machine condition. Moreover, for each available physical magnitude, the reduction of the original number of attributes through the Principal Component Analysis leads to retain a reduced number of significant features that allows achieving the final diagnosis outcome by a multi-label classification tree. The effectiveness of the method was validated by using a complete set of experimental data acquired from a laboratory electromechanical system, where a healthy and seven faulty scenarios were assessed. Also, the interpretation of the results do not require any prior expert knowledge and the robustness of this proposal allows its application in industrial applications, since it may deal with different operating conditions such as different loads and operating frequencies. Finally, the performance was evaluated using multi-label measures, which to the best of our knowledge, is an innovative development in the field condition monitoring and fault identification.

Observation of Metacognitive Skills in Natural Environments: A Longitudinal Study With Mixed Methods.

Recent studies pointing to evaluation methods in natural environments suggest that their use in the analysis of metacognitive skills provides more precise information than the use of off-line evaluation methods. In this research, mixed methods are used over one academic year for the evaluation of the metacognitive skills that students of Secondary Education apply to solve physics problems. The objectives of this study are to analyze the use of metacognitive skills in natural environments and to study behavioral patterns of student learning through a longitudinal study. A total of 509 recordings of think-aloud protocols are analyzed through the categorization of the responses (liquefying) and the protocol of Van der Stel and Veenman for the analysis of the quality of metacognitive skills. Fewer conceptual errors and less uncertainty over vocabulary were noted during the academic year. Nevertheless, a degree of ambiguity persisted in the understanding of physics concepts. The metacognitive skills of Orientation and Planning were used more than any others. The technique of graph analysis is also applied, to establish the patterns of behavior of each student throughout the academic year. Different patterns were found, the analysis of which helped to identify academically challenged and at-risk students. The use of mixed observation techniques and graph analysis facilitated information on the pace of learning of each student. Future studies will be directed at proposals for the automation of these evaluation techniques in natural learning environments.

Differential efficacy of the resources used in b-learning environments / Eficacia diferencial de los recursos empleados en entornos b-learning

Background: Learning is increasingly frequent in B-Learning spaces. It is therefore necessary to study the characteristics that guarantee deeper and more successful learning in these learning environments. Method: We work with sample of 233 university students using the Moodle 3.1 platform in the third year of their degrees in Health Sciences. The effectiveness of four types of B-Learning on Learning Results (LR), behaviors on the platform, and student satisfaction are all studied. Prior knowledge is also used as a covariable. Results: It was found that the B-Learning environment in which the students obtained better general Outcomes Learning Results (LR) and a higher degree of satisfaction was the one that included the use of infographics and virtual laboratories based on Self-Regulated Learning (SRL). Conclusions: The design of B-Learning environments together with the use of SRL, is a factor that enhances effective learning and increases student satisfaction, especially if they include infographics and virtual laboratories. In addition, the use of these resources implements better overall LR on a larger number of students. Likewise, it promotes more homogeneous groups in the general LO. Future investigations will be aimed at verifying these results in other knowledge branches

Antecedentes: es cada vez más frecuente que el aprendizaje se realice en espacios B-Learning. Por ello, es preciso estudiar cuáles son las características que garantizan en estos entornos aprendizajes más profundos y exitosos. Método: se trabajó en la plataforma Moodle 3.1 con una muestra 233 estudiantes universitarios de tercero de grado en la rama de Ciencias de la Salud. Se estudió la efectividad de cuatro tipos de B-Learning sobre los Resultados de Aprendizaje (RA), las conductas de aprendizaje y la satisfacción de los estudiantes. Asimismo, se utilizó como covariable los conocimientos previos. Resultados: se halló que el entorno B-Learning en el que los estudiantes obtuvieron mejores RA generales y mayor grado de satisfacción fue el que incluía el uso de infografías y de laboratorios virtuales basados en aprendizaje autorregulado (SRL). Conclusiones: el diseño de entornos B-Learning, junto con la utilización de SRL, es un factor que potencia aprendizajes eficaces e incrementa la satisfacción de los estudiantes, especialmente si incluyen infografías y laboratorios virtuales. Además, el uso de estos recursos implementa mejores RA generales en un mayor número de estudiantes. Futuras investigaciones irán dirigidas a comprobar estos resultados en otras ramas de conocimiento

Differential efficacy of the resources used in B-learning environments.

BACKGROUND: Learning is increasingly frequent in B-Learning spaces. It is therefore necessary to study the characteristics that guarantee deeper and more successful learning in these learning environments. METHOD: We work with sample of 233 university students using the Moodle 3.1 platform in the third year of their degrees in Health Sciences. The effectiveness of four types of B-Learning on Learning Results (LR), behaviors on the platform, and student satisfaction are all studied. Prior knowledge is also used as a covariable. RESULTS: It was found that the B-Learning environment in which the students obtained better general Outcomes Learning Results (LR) and a higher degree of satisfaction was the one that included the use of infographics and virtual laboratories based on Self-Regulated Learning (SRL). CONCLUSIONS: The design of B-Learning environments together with the use of SRL, is a factor that enhances effective learning and increases student satisfaction, especially if they include infographics and virtual laboratories. In addition, the use of these resources implements better overall LR on a larger number of students. Likewise, it promotes more homogeneous groups in the general LO. Future investigations will be aimed at verifying these results in other knowledge branches.

Does the Use of Learning Management Systems With Hypermedia Mean Improved Student Learning Outcomes?

Learning management systems (LMSs) that incorporate hypermedia Smart Tutoring Systems and personalized student feedback can increase self-regulated learning (SRL), motivation, and effective learning. These systems are studied with the following aims: (1) to verify whether the use of LMS with hypermedia Smart Tutoring Systems improves student learning outcomes; (2) to verify whether the learning outcomes will be grouped into performance clusters (Satisfactory, Good, and Excellent); and (3) to verify whether those clusters will group together the different learning outcomes assessed in four different evaluation procedures. Use of the LMS with hypermedia Smart Tutoring Systems was studied among students of Health Sciences, all of whom had similar test results in the use of metacognitive skills. It explained 38% of the variance in student learning outcomes in the evaluation procedures. Likewise, three clusters that grouped the learning outcomes in relation to the variable 'Use of an LMS with hypermedia Smart Tutoring Systems vs. No use' explained 60.4% of the variance. Each cluster grouped the learning outcomes in the different evaluation procedures. In conclusion, LMS with hypermedia Smart Tutoring Systems in Moodle increased the effectiveness of student learning outcomes, above all in the individual quiz-type tests. It also facilitated personalized learning and respect for the individual pace of student-learning. Hence, modules for the analysis of supervised, unsupervised and multivariate learning should be incorporated into the Moodle platform to provide teaching tools that will undoubtedly contribute to improvements in student learning outcomes. HIGHLIGHTS -Learning management systems (LMS) that incorporate hypermedia Smart Tutoring Systems and personalized student feedback can increase self-regulated learning (SRL).-Learning management systems with hypermedia Smart Tutoring Systems increased the effectiveness of student learning outcome.-The use of an LMS with hypermedia Smart Tutoring Systems vs. No use' explained 60.4% of the variance in student learning outcome.

How Do B-Learning and Learning Patterns Influence Learning Outcomes?

Learning Management System (LMS) platforms provide a wealth of information on the learning patterns of students. Learning Analytics (LA) techniques permit the analysis of the logs or records of the activities of both students and teachers on the on-line platform. The learning patterns differ depending on the type of Blended Learning (B-Learning). In this study, we analyse: (1) whether significant differences exist between the learning outcomes of students and their learning patterns on the platform, depending on the type of B-Learning [Replacement blend (RB) vs. Supplemental blend (SB)]; (2) whether a relation exists between the metacognitive and the motivational strategies (MS) of students, their learning outcomes and their learning patterns on the platform. The 87,065 log records of 129 students (69 in RB and 60 in SB) in the Moodle 3.1 platform were analyzed. The results revealed different learning patterns between students depending on the type of B-Learning (RB vs. SB). We have found that the degree of blend, RB vs. SB, seems to condition student behavior on the platform. Learning patterns in RB environments can predict student learning outcomes. Additionally, in RB environments there is a relationship between the learning patterns and the metacognitive and (MS) of the students.