Explore the vast range of titles available.

Making the knowledge contained in scientific papers machine-readable and formally computable would allow researchers to take full advantage of this information by enabling integration with other knowledge sources to support data analysis and interpretation. Here we describe Biofactoid, a web-based platform that allows scientists to specify networks of interactions between genes, their products, and chemical compounds, and then translates this information into a representation suitable for computational analysis, search and discovery. We also report the results of a pilot study to encourage the wide adoption of Biofactoid by the scientific community.

The massive volume of scientific papers published in journals or deposited in preprint servers each day makes it difficult for scientists to stay on top of their respective areas of study, leading to a state of \"information overload''. This trend is accelerating, with the total number of published papers climbing by ~9% year-over-year for the past several decades; in biomedicine alone, articles are deposited in PubMed at a rate greater than 2 per minute. Therefore, automated methods are necessary to manage the rapidly growing volume of literature and maximize the pace of scientific discovery. The explosive increase in the volume of scientific literature has coincided with a breakneck pace of progress in natural language processing (NLP). The application of NLP to scientific literature (sometimes called \"scholarly document processing'' or SDP) has enabled literature-scale information extraction, precise search across 100s of millions of papers, and automatic summarization. In this dissertation, I study the methods behind these applications to (1) understand failure modes, (2) reduce the dependence on labelled training data, and (3) develop novel and performant solutions. I make several contributions to the burgeoning field of SDP. First, in the area of information extraction (IE), I quantify the poor generalization of an existing state-of-the-art approach and propose three complementary solutions to close the train-test performance gap. I then introduce a novel architecture that leverages pre-trained language models (PLMs) to improve performance while reducing training time and labelled data requirements. Additionally, I demonstrate that a generative approach to IE can match or exceed the performance of existing discriminative methods while being more flexible. Second, I make fundamental contributions to dense text representations and similarity by developing one of the first unsupervised pre-training strategies for learning high-quality sentence and paragraph embeddings. Finally, I name and study a new framework for query-based, multi-document summarization (MDS) and explore its potential for automatic literature review. I release all research artifacts, including trained models and code for reproducing our results to enable future work on these important tasks.

Arbuscular mycorrhizal fungi (AMF) are known to improve plant fitness through the establishment of mycorrhizal symbioses. Genetic and phenotypic variations among closely related AMF isolates can significantly affect plant growth, but the genomic changes underlying this variability are unclear. To address this issue, we improved the genome assembly and gene annotation of the model strain Rhizophagus irregularis DAOM197198, and compared its gene content with five isolates of R. irregularis sampled in the same field. All isolates harbor striking genome variations, with large numbers of isolate-specific genes, gene family expansions, and evidence of interisolate genetic exchange. The observed variability affects all gene ontology terms and PFAM protein domains, as well as putative mycorrhiza-induced small secreted effector-like proteins and other symbiosis differentially expressed genes. High variability is also found in active transposable elements. Overall, these findings indicate a substantial divergence in the functioning capacity of isolates harvested from the same field, and thus their genetic potential for adaptation to biotic and abiotic changes. Our data also provide a first glimpse into the genome diversity that resides within natural populations of these symbionts, and open avenues for future analyses of plant-AMF interactions that link AMF genome variation with plant phenotype and fitness.

Arbuscular mycorrhizal fungi (AMF) are known to improve plant fitness through the establishment of mycorrhizal symbioses. Genetic and phenotypic variations among closely related AMF isolates can significantly affect plant growth, but the genomic changes underlying this variability are unclear. To address this issue, we improved the genome assembly and gene annotation of the model strain Rhizophagus irregularis DAOM197198, and compared its gene content with five isolates of R. irregularis sampled in the same field. All isolates harbor striking genome variations, with large numbers of isolate-specific genes, gene family expansions, and evidence of interisolate genetic exchange. The observed variability affects all gene ontology terms and PFAM protein domains, as well as putative mycorrhiza-induced small secreted effector-like proteins and other symbiosis differentially expressed genes. High variability is also found in active transposable elements. Overall, these findings indicate a substantial divergence in the functioning capacity of isolates harvested from the same field, and thus their genetic potential for adaptation to biotic and abiotic changes. Our data also provide a first glimpse into the genome diversity that resides within natural populations of these symbionts, and open avenues for future analyses of plant–AMF interactions that link AMF genome variation with plant phenotype and fitness.

College degree attainment has been described as a pathway to meritocratic societal outcomes (Seidman, 2005; Torche, 2011). First-generation college students (FGCS) have been documented as historically disenfranchised from college outcomes (Cataldi et al., 2018; Choy, 2001). In addition, career outcomes are equivalent for first- and continuing-generation (CGCS) graduates who obtain a bachelor’s degree (Ford, 2018; Torche, 2011). Data support that FGCS are enrolling in agriculture and the related sciences at higher rates than in other disciplines. Lesser levels of engagement and involvement by FGCS in college experiences have been historically correlated to departure (Dika & D’Amico, 2016; Pascarella et al., 2004). Students classified as FGCS have also been documented to associate with a lower sense of belonging in postsecondary institutions (Soria & Stebleton, 2012). Recent studies have been used to document that FGCS are not less engaged in college, but are differently engaged compared to their CGCS counterparts (Yee, 2016).Astin’s (1977, 1991) I-E-O model and theory of involvement provided the foundation for this research. According to Astin (1977), inputs interact with experiences to shape outcomes. Additionally, Astin (1991) posited that experiences have varying effects on outcomes based on quality and quantity of involvement. The purpose of this study was to describe multiple decades of characteristics and behaviors of students enrolled in colleges of agriculture and the related sciences (ARS) in the United States. Specifically, the study was designed to describe patterns of graduation, and trends in student campus involvement among underrepresented populations including first-generation and continuing-generation students. The objectives were to describe, nationally, across multiple decades, the types of degrees in which students graduated in ARS, as well as the rates of graduation by sex and race. Additionally, further objectives of this study were to describe The Ohio State University (OSU) students’ campus involvement, participation in leadership, and sense of belonging. Parental level of education was used to define student generational status as input categories. The variables of on-campus involvement, student sense of belonging, leadership capacity, efficacy, and motivation were operationalized as experiences. Graduation was the conceptual outcome for students, though this study did not measure outcomes.Secondary datasets were used to describe experiences of students at The Ohio State University (OSU), and to compare generational status and colleges. Responses to the Student Life Survey 2019 cohort (n = 726), and the Multi-institutional Study of Leadership 2018 cohort (n = 2635) were used as secondary datasets to describe student experiences. Descriptive statistics were calculated for the variables of on-campus involvement, leadership capacity, leadership efficacy, leadership motivation, and sense of belonging.Data on graduation trends were compiled from the digitally available Digest of Education Statistics. Descriptive results were used to show that nationally, for both the all degree areas, and the degrees in ARS, quantity of graduates continuously increased in all degrees since 1970. During the last decade, graduates in all degree types for ARS, increased by approximately 41%. Graduates with a bachelor’s degree in ARS comprised less than 2% of all bachelor’s degrees conferred. As of the 1980-81 academic year, bachelor’s degree graduates, in all areas, were predominately women. Graduates with bachelor’s degrees in ARS were mostly women beginning in 2011-12.The quantity of graduates obtaining a bachelor and associate degrees, specifically for ARS, continuously increased over the last decade. The number of degrees conferred, in the last decade, to those classified as Pacific Islander and Alaskan Native or American Indian, decreased. Additionally, the percentage of graduates in ARS became less White, while Hispanic and multi-racial persons doubled their percentage makeup in the last decade.A majority of students at OSU, regardless of generational status, were engaged in student organizations. For students affiliated with the College of Food, Agricultural, and Environmental Sciences (CFAES), the two on-campus activities in which most respondents reported participating were student organizations (69.4%), and on-campus jobs (53.1%). The level of participation in activities, by CFAES students, was greater than that of the overall university. FGCS affiliated with CFAES were found to participate in undergraduate research and capstone experiences more than their continuing-generation peers. Other findings from this study included: more FGCS are not involved in campus activities compared to CGCS for OSU, and CFAES students are reported as more involved compared to overall OSU students.Respondents were either neutral, or they agreed with each of the constructs for leadership, regardless of area of study, or generational-status. It was found that no statistically significant differences existed for the constructs of leadership or sense of belonging as measured by area of study, or generational-status.Colleges of similar size and diversity characteristics compared to CFAES were found to have similar levels of sense of belonging, and leadership indicators. The College of Education and Human Ecology (EHE) was found to be less involved than CFAES, regardless of students’ generational status.Findings and conclusions from this study are not generalizable. Students at OSU affiliated with CFAES, regardless of generational status, participated in student activities more than peers in other colleges. Additionally, there was no difference within the sample, by college major or generational status, for leadership indicators or sense of belonging. The lack of difference in leadership indicators and sense of belonging between FGCS and CGCS should be further explored, as previous literature suggests a disparate effect of generational-status on these indicators (Soria & Stebleton, 2012). Further investigation is needed to explore the potential effects of the higher rate of participation by CFAES FGCS students in undergraduate research on participation in graduate education, which has been documented as an educational opportunity where FGCS do not participate equally (Choy, 2001). Research designed to test the influence of differing experiences on academic outcomes is needed.