Search for: All records

Recognizing the promise of natural language interfaces to databases, prior studies have emphasized the development of text-to-SQL systems. Existing research has generally focused on generating SQL statements from text queries, and the broader challenge lies in inferring new information about the returned data. Our research makes two major contributions to address this gap. First, we introduce a novel Internet-of-Things (IoT) text-to-SQL dataset comprising 10,985 text-SQL pairs and 239,398 rows of network traffic activity. The dataset contains additional query types limited in prior text-to-SQL datasets, notably, temporal-related queries. Our dataset is sourced from a smart building’s IoT ecosystem exploring sensor read and network traffic data. Second, our dataset allows two-stage processing, where the returned data (network traffic) from a generated SQL can be categorized as malicious or not. Our results show that joint training to query and infer information about the data improves overall text-to-SQL performance, nearly matching that of substantially larger models. We also show that current large language models (e.g., GPT3.5) struggle to infer new information about returned data (i.e., they are bad at tabular data understanding), thus our dataset provides a novel test bed for integrating complex domain-specific reasoning into LLMs. 

Abstract Cancer is an intricate disease accountable for the deaths of over 10 million people per year in the United States of America. Several scientific studies showed that the cancer stem cell (CSC) markers have prognostic significance in various cancers and are crucial for designing anticancer drugs to lower cancer death. However, there was a lack of rapid, accurate identification, and analysis, of the prognostic cancer stem cell (CSC) biomarkers in numerous cancer patients. In our laboratory, we identified and analyzed prognostic lung cancer stem cell markers (LCSCs) by using the Immunofluorescence microtissue array (IMA) technique in different lung cancer patient’s tissue biopsy samples and observed that the increased expression of LCSCs principally, CD44 and CD80 in stage IIIA lung cancer tissues compared to normal lung biopsy tissues. We also investigated pancreatic cancer stem cell biomarkers (PAN CSCs) namely CD44 and CD80 with the IMA technique in pancreatic biopsy tissues. The CD44 fluorescence proved an increased expression in adenocarcinoma pancreatic cell tissues when compared to CD80. We also studied and analyzed the stage progression with ovarian cancer stem cell biomarkers (OCSCs) chiefly CD54 and CD44 using the IMA technique in ovarian cancer patients and normal biopsy tissues. The increased expression of CD44 and CD54 were observed in Stage III ovarian cancer tissues compared to normal ovarian tissue indicating the potential role of these OCSC’s biomarkers for the prognosis of ovarian cancer pathogenesis. Our results of prognostic cancer stem cell biomarkers of lung, pancreatic, and ovarian cancers have been analyzed by one-way ANOVA and bioinformatics software (Reactome, Cytoscape PSICQUIC services, STRING) to find underlying molecular mechanism of target gene regulation of increased expression of prognostic CSCs which may give a clue for the prevention and treatment of these cancers. Further research is warranted for these lung, pancreatic, and ovarian CSCs which could be valuable for clinical trials and drug discovery against these CSC biomarkers at early-stage development. Citation Format:Madhumita Das, Kymkecia Henry, Djarie Armstrong, Charle Truman, Charlie Kendrick, Maya S. Saunders, Juan E. Anderson, Malcolm J. Lovett, Rose Stiffin, Ayivi Huisso, Donrie Purcell, Marco Ruiz, Paulo Chaves, Jayanta Kumar Das. Immunofluorescence microtissue array (IMA) for detection of prognostic cancer stem cell biomarkers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 7077.