The COVID-19 pandemic has emphasized the importance of accurate detection of known and emerging pathogens. However, robust characterization of pathogenic sequences remains an open challenge. To address this need we developed SeqScreen, which accurately characterizes short nucleotide sequences using taxonomic and functional labels and a customized set of curated Functions of Sequences of Concern (FunSoCs) specific to microbial pathogenesis. We show our ensemble machine learning model can label protein-coding sequences with FunSoCs with high recall and precision. SeqScreen is a step towards a novel paradigm of functionally informed synthetic DNA screening and pathogen characterization, available for download atwww.gitlab.com/treangenlab/seqscreen.
1 September 2023 in Proceedings of the Human Factors and Ergonomics Society Annual Meeting
Intelligence analysts pore over myriad sources to identify key facts, make sense of them, and use that “sensemaking” to inform mission-critical action. Often these operations are conducted 24/7 with analysts working in shifts. To understand these processes in situ , we conducted interviews with seven former and current United States intelligence analysts. We develop theory and advance a descriptive model of shift handover processes in this domain, which takes into account sensemaking, the communication process, and the features of big data. Our qualitative analysis reveals the following tensions analysts face in shift handovers (1) narrowing versus broadening one’s own sensemaking, (2) sensemaking versus encoding for the subsequent analyst, and (3) trusting versus validating the prior analyst’s sensemaking. In addition, we describe several practical takeaways.
Artificial intelligence (AI) is suggested as a support for virtual collaboration. We conducted a between-subjects experiment ( N = 151) to compare two AI summarizers using the Human-Agent Teaming on Intelligence Tasks (HATIT) platform. Participants reviewed study-created documents, with the HATIT AI summarizer providing either an informative or an indicative summary. Dispositional trust in AI influenced learned trust in the AI summarizer and greater trust developed in the informative condition. Compared to the indicative condition, participants in the informative condition also displayed some evidence of faster but less frequent attention to newly arrived teammate information. Problem-solving accuracy was comparable across conditions.
In complex domains such as intelligence analysis and healthcare, teams often need to continuously work toward long-term goals. These scenarios involve sequential work, with teammates performing their roles in shifts. Existing teamwork research acknowledges the myriad factors that influence how information is cognitively processed and collaboratively deliberated to arrive at a shared decision. Yet, much of this research describes synchronous and/or reciprocal interdependence rather than sequential interdependence. This paper presents a descriptive model of collective judgment formation (CJF) in sequential teams. CJF involves the development of judgments by drawing on information from diverse sources, including both humans and AI, such that a mosaic comes into form even as pieces appear and others change shape. CJF also often entails situations with an overwhelming variety, volume, and velocity of information. Our model can be applied to numerous sequential scenarios that incorporate shift handovers and processing data into intelligence.