Documentation Matters: Human-Centered AI System to Assist Data Science Code Documentation in Computational Notebooks

Computational notebooks allow data scientists to express their ideas through a combination of code and documentation. However, data scientists often pay attention only to the code, and neglect creating or updating their documentation during quick iterations. Inspired by human documentation practices learned from 80 highly-voted Kaggle notebooks, we design and implement Themisto, an automated documentation generation system to explore how human-centered AI systems can support human data scientists in the machine learning code documentation scenario. Themisto facilitates the creation of documentation via three approaches: a deep-learning-based approach to generate documentation for source code, a query-based approach to retrieve online API documentation for source code, and a user prompt approach to nudge users to write documentation. We evaluated Themisto in a within-subjects experiment with 24 data science practitioners, and found that automated documentation generation techniques reduced the time for writing documentation, reminded participants to document code they would have ignored, and improved participants’ satisfaction with their computational notebook.

Quality assessment of anatomical MRI images from Generative Adversarial Networks: human assessment and image quality metrics

Background: Generative Adversarial Networks (GANs) can synthesize brain images from image or noise input. So far, the gold standard for assessing the quality of the generated images has been human expert ratings. However, due to limitations of human assessment in terms of cost, scalability, and the limited sensitivity of the human eye to more subtle statistical relationships, a more automated approach towards evaluating GANs is required. New method: We investigated to what extent visual quality can be assessed using image quality metrics and we used group analysis and spatial independent components analysis to verify that the GAN reproduces multivariate statistical relationships found in real data. Reference human data was obtained by recruiting neuroimaging experts to assess real Magnetic Resonance (MR) images and images generated by a Wasserstein GAN. Image quality was manipulated by exporting images at different stages of GAN training. Results: Experts were sensitive to changes in image quality as evidenced by ratings and reaction times, and the generated images reproduced group effects (age, gender) and spatial correlations moderately well. We also surveyed a number of image quality metrics which consistently failed to fully reproduce human data. While the metrics Structural Similarity Index Measure (SSIM) and Naturalness Image Quality Evaluator (NIQE) showed good overall agreement with human assessment for lower-quality images (i.e. images from early stages of GAN training), only a Deep Quality Assessment (QA) model trained on human ratings was sensitive to the subtle differences between higher-quality images. Conclusions: We recommend a combination of group analyses, spatial correlation analyses, and both distortion metrics (SSIM, NIQE) and perceptual models (Deep QA) for a comprehensive evaluation and comparison of brain images produced by GANs.

A Joint Use of Pooling And Imputation For Genotyping SNPs

Background: Despite continuing technological advances, the cost for large-scale genotyping of a high number of samples can be prohibitive. The purpose of this study is to design a cost-saving strategy for SNP genotyping. We suggest making use of pooling, a group testing technique, to drop the amount of SNP arrays needed. We believe that this will be of the greatest importance for non-model organisms with more limited resources in terms of cost-efficient large-scale chips and high-quality reference genomes, such as application in wildlife monitoring, plant and animal breeding, but it is in essence species-agnostic. The proposed approach consists in grouping and mixing individual DNA samples into pools before testing these pools on bead-chips, such that the number of pools is less than the number of individual samples. We present a statistical estimation algorithm, based on the pooling outcomes, for inferring marker-wise the most likely genotype of every sample in each pool. Finally, we input these estimated genotypes into existing imputation algorithms. We compare the imputation performance from pooled data with the Beagle algorithm, and a local likelihood-aware phasing algorithm closely modeled on MaCH that we implemented. Results: We conduct simulations based on human data from the 1000 Genomes Project, to aid comparison with other imputation studies. Based on the simulated data, we find that pooling impacts the genotype frequencies of the directly identifiable markers, without imputation. We also demonstrate how a combinatorial estimation of the genotype probabilities from the pooling design can improve the prediction performance of imputation models. Our algorithm achieves 93% concordance in predicting unassayed markers from pooled data, thus it outperforms the Beagle imputation model which reaches 80% concordance. We observe that the pooling design gives higher concordance for the rare variants than traditional low-density to high-density imputation commonly used for cost-effective genotyping of large cohorts. Conclusions: We present promising results for combining a pooling scheme for SNP genotyping with computational genotype imputation, as demonstrated in simulations on human data, while using half the number of assays needed for sample-wise genotyping. These results could find potential applications in any context where the genotyping costs form a limiting factor on the study size, such as in marker-assisted selection in plant breeding.

The Interplay Between Stress Related Genes and Its Role in Human Longevity: Insights for Translational Studies

Human lifespan is a multifactorial trait resulted from complicated interplay among many genetic and environmental factors. Despite substantial progress in clarifying many aspects of lifespan’ variability the mechanism of its multifactorial regulation remains unclear. In this paper we investigate the role of genes from integrated stress response (ISR) pathway in such regulation. Experimental studies showed that persistent cellular stress may result in cellular senescence (for proliferating cells), or in apoptosis (for post-mitotic cells) which may affect health and lifespan in laboratory animals. These studies also showed which ISR genes are likely to interplay to produce joint effects on these traits. Note that in humans, the interplay between these genes does not necessarily influence these traits. This is because biological mechanisms regulating these traits in laboratory animals and humans may differ. This means that, when possible, the experimentally detected connections promising for human applications, should be verified using available human data before their testing in expensive clinical trials. In this paper we used HRS data to test connection between SNPs from the EIF2AK4 gene that senses cellular stress signals and the DDIT3 gene from the apoptosis regulation part of the ISR. We found genome wide significant associations between interacting SNPs from these genes and longevity. This result shows that available human data may be successfully used for making important steps in translation of experimental research findings towards their application in humans. Following this strategy may increase efficiency of clinical trials aiming to find appropriate medications to promote human health and longevity.

A Temporal Model of Aural Frequency Discrimination

<p>The importance of temporal information versus place information in frequency analysis by the ear is a continuing controversy. This dissertation developes a temporal model which simulates human frequency discrimination. The model gives guantitative measures of performance for the discrimination of sinusoids in white gaussian noise. The model simulates human frequency discrimination performance as a function of frequency and signal-to-noise ratio. The model's predictions are based on the temporal intervals between the positive axis crossings of the stimulus. The histograms of these temporal intervals were used as the underlying distributions from which indices of discriminability were calculated. Human freguency discrimination data was obtained for five observers as a function of frequency and signal-to-noise ratio. The data were analysed using the method of Group-operating-characteristic (GOC) Analysis. This method of analysis statistically removes unique noise from data. The unique noise was removed by summing observers' ratings for identical stimuli. This method of analysis gave human frequency discrimination data with less unigue noise than any existing frequency data. The human data were used for evaluating the model. The GOC Analysis was also used to study the improvement in d' as a function of stimulus replications and signal-to-noise ratio. The model was a good fit to the human data at 250 Hz, for two signal-to-noise ratios. The model did not fit the data at 1000 Hz or 5000 Hz. There was some evidence of a transition occuring at 1000 Hz. This investigation supported the idea that human frequency discrimination relies on a temporal mechanism at low frequencies with a transition to some other mechanism at about lO00 Hz.</p>

A Temporal Model of Aural Frequency Discrimination

<p>The importance of temporal information versus place information in frequency analysis by the ear is a continuing controversy. This dissertation developes a temporal model which simulates human frequency discrimination. The model gives guantitative measures of performance for the discrimination of sinusoids in white gaussian noise. The model simulates human frequency discrimination performance as a function of frequency and signal-to-noise ratio. The model's predictions are based on the temporal intervals between the positive axis crossings of the stimulus. The histograms of these temporal intervals were used as the underlying distributions from which indices of discriminability were calculated. Human freguency discrimination data was obtained for five observers as a function of frequency and signal-to-noise ratio. The data were analysed using the method of Group-operating-characteristic (GOC) Analysis. This method of analysis statistically removes unique noise from data. The unique noise was removed by summing observers' ratings for identical stimuli. This method of analysis gave human frequency discrimination data with less unigue noise than any existing frequency data. The human data were used for evaluating the model. The GOC Analysis was also used to study the improvement in d' as a function of stimulus replications and signal-to-noise ratio. The model was a good fit to the human data at 250 Hz, for two signal-to-noise ratios. The model did not fit the data at 1000 Hz or 5000 Hz. There was some evidence of a transition occuring at 1000 Hz. This investigation supported the idea that human frequency discrimination relies on a temporal mechanism at low frequencies with a transition to some other mechanism at about lO00 Hz.</p>

Strategi Ekstrakurikuler Muhadharah dalam Melatih Kemampuan Public Speaking Siswa Mi

This study aims to determine the implementation of Muhadharah in training public speaking skills at MI Ruhul Amin, Muhadharah's extracurricular strategies in training public speaking skills, and its inhibiting and supporting factors. This type of research was conducted in the form of a qualitative descriptive field study. This study's sources of data include principals, teachers, students, and non-human data sources consisting of relevant madrasah documents and data. The results showed that the extracurricular activities of Muhadharah at MI Ruhul Amin were held every Saturday afternoon. The strategy used is to make a muhadharah schedule, compile and correct the speech's text, take turns choosing a place for muhadharah, take part in competitions or competitions. Inhibiting factors include students' lack of interest in muhadharah practice, lack of confidence, incomplete facilities, students' different character, and monotonous material delivery. These supporting factors include the extracurricular muhadharah that is carried out regularly, the existence of learning evaluations, the existence of sanctions for students who violate the rules, and quality supervisors