ABSTRACT
We seek to transform how new and emergent variants of pandemiccausing viruses, specifically SARS-CoV-2, are identified and classified. By adapting large language models (LLMs) for genomic data, we build genome-scale language models (GenSLMs) which can learn the evolutionary landscape of SARS-CoV-2 genomes. By pretraining on over 110 million prokaryotic gene sequences and finetuning a SARS-CoV-2-specific model on 1.5 million genomes, we show that GenSLMs can accurately and rapidly identify variants of concern. Thus, to our knowledge, GenSLMs represents one of the first whole genome scale foundation models which can generalize to other prediction tasks. We demonstrate scaling of GenSLMs on GPU-based supercomputers and AI-hardware accelerators utilizing 1.63 Zettaflops in training runs with a sustained performance of 121 PFLOPS in mixed precision and peak of 850 PFLOPS. We present initial scientific insights from examining GenSLMs in tracking evolutionary dynamics of SARS-CoV-2, paving the path to realizing this on large biological data.
ABSTRACT
This article inquires into whether and how uncertainty reduction theory and problematic integration theory, two theories relevant to social support as enacted within a chapter of the Man-to-Man prostate cancer support group, inform us of how such groups can assist group members most effectively. Interview data from members of a prostate cancer support group shed light on theoretical assumptions about uncertainty. Although the group applies elements of both theories, prostate cancer survivors likely would benefit from a more comprehensive and flexible treatment of social support theory.
Subject(s)
Communication , Health Education , Health Status , Professional-Patient Relations , Prostatic Neoplasms/psychology , Psychological Theory , Social Support , Aged , Anxiety Disorders/diagnosis , Anxiety Disorders/epidemiology , Anxiety Disorders/etiology , Attitude to Health , Humans , Male , Prostatic Neoplasms/epidemiologyABSTRACT
Extending literature on health information to entertainment television, we analyze the prostate cancer narrative presented in the police drama, NYPD Blue. We explain how the physician-patient interaction depicted on the show followed (and sometimes did not follow) the medical dialogue model. Findings reveal that the producers of this show advocate a more dialogic model of medical interaction. Portrayals of incompetent, ineffective physicians are contrasted with the superior, effective efforts of other physicians. The audience learns that a non-dialogic approach characterizes "bad doctors," while the dialogic method typifies "good doctors." Likewise, medical professionals can use such texts to enhance physician-patient interaction.
Subject(s)
Clinical Competence , Communication , Physician-Patient Relations , Prostatic Neoplasms , Television , Drama , Humans , Male , PoliceABSTRACT
Prostate cancer affects men and their loved ones; consequently, survivors and their wives can gain from social support throughout the illness experience. After observing meetings of a support group for prostate cancer survivors and their partners, the authors used the constant comparison method to draw conclusions about the types of support generated in the men's and women's divisions of the group. The authors concluded that both divisions served as sites of information but not as scenes of practical assistance. The authors also found that the discursive practices of the groups and the structural elements of the group meetings inhibited emotional support through topic turning, comparisons between members, and the role of group facilitators. The authors consider the study's implications for support group leaders and scholars.
