Publications

Microbes possess a natural capacity to resist antimicrobial agents (substances or compounds that can stop or slow down their growth). Due to bad administration of antibiotics uses, dearth of new development pipelines for antibiotics, poor monitoring and surveillance, lack of decision support systems warranting the right antibiotic administration, and other factors, antimicrobial resistance (AMR) is a global threat that will impact deaths, quality of life postinfection, and economic losses. Artificial Intelligence (AI) can be an emerging tool to fight AMR in different areas ranging from identifying or developing new antibiotics, slowing down AMR development due to better management guided by Machine Learning (ML) models, development of diagnostics arrangements, and supporting clinical decision systems. In this chapter, we present the applications of AI in different areas of early research, clinical setting, and decision-making for health care. Further, we highlight the potential of AI in omics studies for AMR research and diagnosis of AMR and the recent developments within these areas. The adoption of AI is strongly challenged by high accuracy requirements within the healthcare system and hence for better adoption. Therefore, we discussed the impending challenges, upon resolution of which can collectively slow down AMR development. Lastly, we also discuss the recommendations for investing focus within the current AI regime and future developments that can benefit the field.

Targeting EGFR has been effective in RAS/RAF wild-type colorectal cancer (CRC) patients. However, residual tumor relapses, necessitating the importance of biomarker-guided novel therapeutics. We show elevated DKC1 in ∼88% of CRC patients with poor recurrence-free survival. Clinically, DKC1-positive patients exhibit similarity with CMS2 class, the canonical subtype with active WNT signaling. We show functional significance of DKC1 in cell proliferation, stemness, DNA repair, and survival. Further, mice bearing DKC1 knockdown xenografts show ∼81% reduction in tumor burden. Mechanistically, WNT/β-catenin signaling orchestrates DKC1 expression, then, DKC1/SOX2 complex regulates SGPP2, modulating sphingolipids metabolism. Downregulation of DKC1 in CRC lead to reduced SGPP2 levels leading to dysregulation of sphingolipid biosynthesis. Of note, DKC1-high CRC patients show accumulation of ceramides, namely C23 and C24, signifying their utility in diagnosis. Collectively, we delineate the mechanistic circuitry involved in DKC1-mediated CRC progression, propose ceramides as biomarker, and underscore WNT-based therapeutics for DKC1-positive patients.

PARP inhibitors (PARPi) have emerged as a promising targeted therapeutic intervention for metastatic castrate-resistant prostate cancer (mCRPC). However, the clinical utility of PARPi is limited to a subset of patients who harbor aberrations in the genes associated with the homologous recombination (HR) pathway. Here, we report that targeting metastasis-associated lung adenocarcinoma transcript 1 (MALAT1), an oncogenic long noncoding RNA (lncRNA), contrives a BRCAness-like phenotype, and augments sensitivity to PARPi. Mechanistically, we show that MALAT1 silencing reprograms the homologous recombination (HR) transcriptome and makes prostate cancer cells more vulnerable to PARPi. Particularly, coinhibition of MALAT1 and PARP1 exhibits a decline in clonogenic survival, delays resolution of γH2AX foci, and reduces tumor burden in mice xenograft model. Moreover, we show that miR-421, a tumor suppressor miRNA, negatively regulates the expression of HR genes, while in aggressive prostate cancer cases, miR-421 is sequestered by MALAT1, leading to increased expression of HR genes. Conclusively, our findings suggest that MALAT1 ablation confers sensitivity to PARPi, thus highlighting an alternative therapeutic strategy for patients with castration-resistant prostate cancer (CRPC), irrespective of the alterations in HR genes. Significance: PARPi are clinically approved for patients with metastatic CRPC carrying mutations in HR genes, but are ineffective for HR-proficient prostate cancer. Herein, we show that oncogenic lncRNA, MALAT1 is frequently overexpressed in advanced stage prostate cancer and plays a crucial role in maintaining genomic integrity. Importantly, we propose a novel therapeutic strategy that emphasizes MALAT1 inhibition, leading to HR dysfunction in both HR-deficient and -proficient prostate cancer, consequently augmenting their susceptibility to PARPi.