In summer 2024, as part of Arizona State University’s collaboration with OpenAI, the ASU Library launched a pilot project using the AI tool ChatGPT. This project aims to enhance the discoverability and curation of digital collections within the library’s repository ecosystem. The use of AI in libraries is gaining attention, with many institutions exploring AI for generating descriptive metadata. ASU Library’s extensive repository platforms, including an institutional repository, data repository, and a digital collections platform, hold approximately over 10,000 objects, with numbers expected to grow. The library lacks a dedicated position for creating metadata, with the responsibility distributed among various units already tasked with other duties. This project aims to determine whether ChatGPT can effectively generate accurate metadata that meets best practices. The library will use an existing archival collection of government documents, which already has human-created metadata, as a benchmark, in comparing the generated metadata for the fields Title, Description, and Keywords. By comparing ChatGPT-generated metadata to the existing metadata, the library will assess the relevance of AI outputs and the level of oversight required. If the AI-generated metadata shows minimal variance from the human-created metadata, the workflow could expand to other collections and reduce the backlog of unpublished archival collections that require descriptive metadata.

Acute Myeloid Leukemia (AML) is a fatal hematological cancer. The genetic abnormalities underlying AML are extremely heterogeneous among patients, making prognosis and treatment selection very difficult. While clinical proteomics data has the potential to improve prognosis accuracy, thus far, the quantitative means to do so have yet to be developed. Here we report the results and insights gained from the DREAM 9 Acute Myeloid Prediction Outcome Prediction Challenge (AML-OPC), a crowdsourcing effort designed to promote the development of quantitative methods for AML prognosis prediction. We identify the most accurate and robust models in predicting patient response to therapy, remission duration, and overall survival. We further investigate patient response to therapy, a clinically actionable prediction, and find that patients that are classified as resistant to therapy are harder to predict than responsive patients across the 31 models submitted to the challenge. The top two performing models, which held a high sensitivity to these patients, substantially utilized the proteomics data to make predictions. Using these models, we also identify which signaling proteins were useful in predicting patient therapeutic response.