Providing personalized, detailed feedback at scale in large undergraduate STEM courses remains a persistent challenge. We present an empirically evaluated practice exam system that integrates AI generated feedback with targeted textbook references, deployed in a large introductory biology course. Our system specifically aims to encourage metacognitive behavior by asking students to explain their answers and declare their confidence. It uses OpenAI's GPT-4o to generate personalized feedback based on this information, while directing them to relevant textbook sections. Through detailed interaction logs from consenting participants across three midterms (541, 342, and 413 students respectively), totaling 28,313 question-student interactions across 146 learning objectives, along with 279 post-exam surveys and 23 semi-structured interviews, we examined the system's impact on learning outcomes and student engagement. Analysis showed that across all midterms, the different feedback types showed no statistically significant differences in performance, though there were some trends suggesting potential benefits worth further investigation. The system's most substantial impact emerged through its required confidence ratings and explanations, which students reported transferring to their actual exam strategies. Approximately 40% of students engaged with textbook references when prompted by feedback---significantly higher than traditional reading compliance rates. Survey data revealed high student satisfaction (M=4.1/5), with 82.1% reporting increased confidence on midterm topics they had practiced, and 73.4% indicating they could recall and apply specific concepts from practice sessions. Our findings demonstrate how thoughtfully designed AI-enhanced systems can scale formative assessment while promoting sustainable study practices and self-regulated learning behaviors, suggesting that embedding structured reflection requirements may be more impactful than sophisticated feedback mechanisms.

A new approach to online discussion, which situates student discussions in the margins of the course content, can enhance student engagement with course materials. However, in high-enrollment classes, the large number of comments can overwhelm and intimidate students. Some become frustrated by the volume of potential online interactions and by a perceived lack of immediate relevance to their studies. Likewise, instructors are disappointed when outstanding discussions, that they deem valuable for all to see, get lost in the clutter. To address these challenges, we propose visual spotlighting mechanisms for increasing the saliency of selected comments. We piloted and deployed multiple designs in two high-enrollment biology courses at a large public university in the United States. Interviews, surveys, and a controlled experiment show that spotlighting relevant comments in heavily annotated texts positively affects students' engagement, measured in terms of their attention to comments, and their reported sense of validation and pride. Students also reported their preferences for certain spotlighting designs.

Online forums are an integral part of modern day courses, but motivating students to participate in educationally beneficial discussions can be challenging. Our proposed solution is to initialize (or “seed”) a new course forum with comments from past instances of the same course that are intended to trigger discussion that is beneficial to learning. In this work, we develop methods for selecting high-quality seeds and evaluate their impact over one course instance of a 186-student biology class. We designed a scale for measuring the “seeding suitability” score of a given thread (an opening comment and its ensuing discussion). We then constructed a supervised machine learning (ML) model for predicting the seeding suitability score of a given thread. This model was evaluated in two ways: first, by comparing its performance to the expert opinion of the course instructors on test/holdout data; and second, by embedding it in a live course, where it was actively used to facilitate seeding by the course instructors. For each reading assignment in the course, we presented a ranked list of seeding recommendations to the course instructors, who could review the list and filter out seeds with inconsistent or malformed content. We then ran a randomized controlled study, in which one group of students was shown seeds that were recommended by the ML model, and another group was shown seeds that were recommended by an alternative model that ranked seeds purely by the length of discussion that was generated in previous course instances. We found that the group of students that received posts from either seeding model generated more discussion than a control group in the course that did not get seeded posts. Furthermore, students who received seeds selected by the ML-based model showed higher levels of engagement, as well as greater learning gains, than those who received seeds ranked by length of discussion.

Students' confusion is a barrier for learning, contributing to loss of motivation and to disengagement with course materials. However, detecting students' confusion in large-scale courses is both time and resource intensive. This paper provides a new approach for confusion detection in online forums that is based on harnessing the power of students' self-reported affective states (reported using a set of pre-defined hashtags). It presents a rule for labeling confusion, based on students' hashtags in their posts, that is shown to align with teachers' judgement. We use this labeling rule to inform the design of an automated classifier for confusion detection for the case when there are no self-reported hashtags present in the test set. We demonstrate this approach in a large scale Biology course using the Nota Bene annotation platform. This work lays the foundation to empower teachers with better support tools for detecting and alleviating confusion in online courses.

Reading material has been part of course teaching for centuries, but until recently students' engagement with that reading, and its effect on their learning, has been difficult for teachers to assess. In this article, we explore the idea of examining cognitive engagement---a measure of how deeply a student is thinking about course material, which has been shown to correlate with learning gains---as it varies over different sections of the course reading material. We show that a combination of automatic classification and visualization of cognitive engagement anchored in the text can give teachers---and not only researchers---valuable insight into their students' thinking, suggesting ways to modify their lectures and their course readings to improve learning. We demonstrate this approach with analyzing students' comments in two different courses (Physics and Biology) using the Nota Bene annotation platform.

Determining affective states such as confusion from students' participation in online discussion forums can be useful for instructors of a large classroom. However, manual annotation of forum posts by instructors or paid crowd workers is both time-consuming and expensive. In this work, we harness affordances prevalent in social media to allow students to self-annotate their discussion posts with a set of hashtags and emojis, a process that is fast and cheap. For students, self-annotation with hashtags and emojis provides another channel for self-expression, as well as a way to signal to instructors and other students on the lookout for certain types of messages. This method also provides an easy way to acquire a labeled dataset of affective states, allowing us distinguish between more nuanced emotions such as confusion and curiosity. From a dataset of over 25,000 discussion posts from two courses containing self-annotated posts by students, we demonstrate how we can identify linguistic differences between posts expressing confusion versus curiosity, achieving 83% accuracy at distinguishing between the two affective states.

BACKGROUND: Cross-species comparisons of gene neighborhoods (also called genomic contexts) in microbes may provide insight into determining functionally related or co-regulated sets of genes, suggest annotations of previously un-annotated genes, and help to identify horizontal gene transfer events across microbial species. Existing tools to investigate genomic contexts, however, lack features for dynamically comparing and exploring genomic regions from multiple species. As DNA sequencing technologies improve and the number of whole sequenced microbial genomes increases, a user-friendly genome context comparison platform designed for use by a broad range of users promises to satisfy a growing need in the biological community. RESULTS: Here we present JContextExplorer: a tool that organizes genomic contexts into branching diagrams. We implement several alternative context-comparison and tree rendering algorithms, and allow for easy transitioning between different clustering algorithms. To facilitate genomic context analysis, our tool implements GUI features, such as text search filtering, point-and-click interrogation of individual contexts, and genomic visualization via a multi-genome browser. We demonstrate a use case of our tool by attempting to resolve annotation ambiguities between two highly homologous yet functionally distinct genes in a set of 22 alpha and gamma proteobacteria. CONCLUSIONS: JContextExplorer should enable a broad range of users to analyze and explore genomic contexts. The program has been tested on Windows, Mac, and Linux operating systems, and is implemented both as an executable JAR file and java WebStart. Program executables, source code, and documentation is available at http://www.bme.ucdavis.edu/facciotti/resources_data/software/.

Nucleosomes, the basic repeat units of eukaryotic chromatin, have been suggested to influence the evolution of eukaryotic genomes, both by altering the propensity of DNA to mutate and by selection acting to maintain or exclude nucleosomes in particular locations. Contrary to the popular idea that nucleosomes are unique to eukaryotes, histone proteins have also been discovered in some archaeal genomes. Archaeal nucleosomes, however, are quite unlike their eukaryotic counterparts in many respects, including their assembly into tetramers (rather than octamers) from histone proteins that lack N- and C-terminal tails. Here, we show that despite these fundamental differences the association between nucleosome footprints and sequence evolution is strikingly conserved between humans and the model archaeon Haloferax volcanii. In light of this finding we examine whether selection or mutation can explain concordant substitution patterns in the two kingdoms. Unexpectedly, we find that neither the mutation nor the selection model are sufficient to explain the observed association between nucleosomes and sequence divergence. Instead, we demonstrate that nucleosome-associated substitution patterns are more consistent with a third model where sequence divergence results in frequent repositioning of nucleosomes during evolution. Indeed, we show that nucleosome repositioning is both necessary and largely sufficient to explain the association between current nucleosome positions and biased substitution patterns. This finding highlights the importance of considering the direction of causality between genetic and epigenetic change.

BACKGROUND: Discovery of functionally significant short, statistically overrepresented subsequence patterns (motifs) in a set of sequences is a challenging problem in bioinformatics. Oftentimes, not all sequences in the set contain a motif. These non-motif-containing sequences complicate the algorithmic discovery of motifs. Filtering the non-motif-containing sequences from the larger set of sequences while simultaneously determining the identity of the motif is, therefore, desirable and a non-trivial problem in motif discovery research. RESULTS: We describe MotifCatcher, a framework that extends the sensitivity of existing motif-finding tools by employing random sampling to effectively remove non-motif-containing sequences from the motif search. We developed two implementations of our algorithm; each built around a commonly used motif-finding tool, and applied our algorithm to three diverse chromatin immunoprecipitation (ChIP) data sets. In each case, the motif finder with the MotifCatcher extension demonstrated improved sensitivity over the motif finder alone. Our approach organizes candidate functionally significant discovered motifs into a tree, which allowed us to make additional insights. In all cases, we were able to support our findings with experimental work from the literature. CONCLUSIONS: Our framework demonstrates that additional processing at the sequence entry level can significantly improve the performance of existing motif-finding tools. For each biological data set tested, we were able to propose novel biological hypotheses supported by experimental work from the literature. Specifically, in Escherichia coli, we suggested binding site motifs for 6 non-traditional LexA protein binding sites; in Saccharomyces cerevisiae, we hypothesize 2 disparate mechanisms for novel binding sites of the Cse4p protein; and in Halobacterium sp. NRC-1, we discoverd subtle differences in a general transcription factor (GTF) binding site motif across several data sets. We suggest that small differences in our discovered motif could confer specificity for one or more homologous GTF proteins. We offer a free implementation of the MotifCatcher software package at http://www.bme.ucdavis.edu/facciotti/resources_data/software/.

SUMMARY: High-throughput DNA sequencing technologies have spurred the development of numerous novel methods for genome assembly. With few exceptions, these algorithms are heuristic and require one or more parameters to be manually set by the user. One approach to parameter tuning involves assembling data from an organism with an available high-quality reference genome, and measuring assembly accuracy using some metrics. We developed a system to measure assembly quality under several scoring metrics, and to compare assembly quality across a variety of assemblers, sequence data types, and parameter choices. When used in conjunction with training data such as a high-quality reference genome and sequence reads from the same organism, our program can be used to manually identify an optimal sequencing and assembly strategy for de novo sequencing of related organisms. AVAILABILITY: GPL source code and a usage tutorial is at http://ngopt.googlecode.com CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data is available at Bioinformatics online.

MOTIVATION: Chromatin immunoprecipitation followed by hybridization to a genomic tiling microarray (ChIP-chip) is a routinely used protocol for localizing the genomic targets of DNA-binding proteins. The resolution to which binding sites in this assay can be identified is commonly considered to be limited by two factors: (1) the resolution at which the genomic targets are tiled in the microarray and (2) the large and variable lengths of the immunoprecipitated DNA fragments. RESULTS: We have developed a generative model of binding sites in ChIP-chip data and an approach, MeDiChI, for efficiently and robustly learning that model from diverse data sets. We have evaluated MeDiChI's performance using simulated data, as well as on several diverse ChIP-chip data sets collected on widely different tiling array platforms for two different organisms (Saccharomyces cerevisiae and Halobacterium salinarium NRC-1). We find that MeDiChI accurately predicts binding locations to a resolution greater than that of the probe spacing, even for overlapping peaks, and can increase the effective resolution of tiling array data by a factor of 5x or better. Moreover, the method's performance on simulated data provides insights into effectively optimizing the experimental design for increased binding site localization accuracy and efficacy. AVAILABILITY: MeDiChI is available as an open-source R package, including all data, from http://baliga.systemsbiology.net/medichi.