What Can Genomics Tell Us About Drug Resistance in India? Insights From Recently Published Research Paper by Professor Shradha Karve

What Does this Research Paper Talk About?

Antimicrobial resistance (AMR) – the ability of bacteria and other microbes to survive and grow in the presence of antibiotics that would have otherwise killed them – is one of the most pressing public health crises of our time. When bacteria develop resistance to the drugs designed to kill them, common infections become life-threatening, routine surgeries become risky, and medical care becomes harder to deliver.

The problem is especially acute in low- and middle-income countries (LMICs) like India, where the burden of infectious disease is high, and surveillance resources are limited. Most AMR monitoring in India has traditionally relied on antibiotic susceptibility testing (AST) – a lab method that checks whether a bacterium is killed by specific antibiotics. While useful, this approach does not reveal which genes underlying resistance, how those genes spread between bacteria, or how India’s resistance landscape compares to global trends. Without the understanding of this genomic layer, our ability to design better diagnostics and treatments remains constrained.

What Did the Study Aim to Find?

This study was motivated by a critical gap: the lack of large-scale datasets in India that link AST with whole-genome sequencing (WGS) data across multiple bacterial species. Professor Karve and her lab set out to build such a dataset – one that would allow us to map the genomic underpinnings of AMR in Indian clinical settings, test how well genomic tools predict resistance compared to standard lab tests, and understand how resistance genes are spread and shared among bacteria.

The ultimate goal, along with knowledge building, is to support the development of faster, more accurate molecular diagnostic tools for AMR in India- tools that could one day replace or supplement the slow, culture-based methods currently in use, particularly in critical care settings like ICUs, where rapid diagnosis can be life-saving.

The Strategy

Between July 2022 and July 2024, the team collected 266 bacterial isolates from two groups of tertiary healthcare hospitals, one in Northern India (New Delhi) and one in Western India (Pune). The isolates represented priority pathogens, including Klebsiella pneumoniae, Escherichia coli, Acinetobacter baumannii, Pseudomonas aeruginosa, Staphylococcus aureus, and Enterococcus, as well as a few other emerging threats.

The researchers selected isolates belonging to high-concern resistance categories: carbapenem-resistant (CR), carbapenem-and-colistin-resistant (CR-Col), extended-spectrum beta-lactam-resistant (ESBL), methicillin-resistant (MR), and vancomycin-resistant (VR). For each isolate, two parallel analyses were carried out:

1. AST is the gold-standard laboratory method that tests how bacteria respond to antibiotics.

2. WGS, which allows us to read the complete DNA of each bacterial isolate and identify resistance genes (ARGs), their locations (chromosomes or plasmids), and the mobile genetic elements that can transfer them between bacteria.

Professor Shradha Karve’s Lab used standard bioinformatic tools including the Resistance Gene Identifier (RGI) linked to the Comprehensive Antibiotic Resistance Database (CARD) for genomic resistance prediction, and MOBsuite and MobileElementFinder for plasmid and mobile element analysis.

Results and Conclusion

The study produced several important findings relevant to how AMR is understood and tracked in India.

Professor Shradha Karve and her team found that genomic tools tend to over-predict resistance. Compared genomic resistance predictions against actual AST results across more than 5000 drug-pathogen combinations, the team found significant discrepancies. The most common pattern was the genomic tool predicting resistance when the bacterium was actually sensitive in the lab and was found in more than 400 cases. This ‘over-prediction’ likely reflects the tool flagging genes that confer only a moderate increase in resistance, not enough to breach clinical treatment thresholds. Crucially, these false positives highlight important gaps that need to be addressed before genomic tools can be reliably used for patient diagnostics.

A rich and diverse resistance gene landscape has also been discovered. Over 80 distinct beta-lactamase genes (enzymes that break down beta-lactam antibiotics) were detected. The NDM (New Delhi Metallo-beta-lactamase) family – genes that confer resistance to carbapenem antibiotics, the last-resort drugs – were the most prevalent. Notably, blaNDM-5 was far more common than blaNDM-1 among E. coli and K. pneumoniae isolates, a finding that contrasts with some prior Indian studies and underscores the evolving resistance landscape.

A significant proportion of resistance genes were found on plasmids. Plasmids are small, circular pieces of DNA that bacteria can share with each other, enabling rapid, horizontal spread of resistance. K. pneumoniae showed the highest burden of plasmid-associated ARGs, highlighting its role as a key vehicle for resistance gene dissemination. Mobile genetic elements like insertion sequences and transposons were also found associated with ARGs, further facilitating spread.

Professor Shradha Karve’s Lab discovered several well-known pathogenic lineages (such as E. coli ST131 and K. pneumoniae ST147), but also noted the diversity and regional specificity of Indian lineages, reinforcing that global AMR data cannot simply be applied to the Indian context.

Impact and Benefits of the Study (Bird-Eye-View)

This study represents the first systematic, multi-species genomic AMR surveillance effort in India to directly compare genomic predictions with phenotypic resistance data across a wide range of drug-pathogen combinations. The findings have several important implications.

• By identifying which drug-pathogen combinations are well-predicted by genomics and which are not, this work provides a roadmap for improving the accuracy of molecular diagnostic tools for AMR. Rapid genomic diagnostics could transform care in ICUs, where delays in identifying the right antibiotic can cost lives.
• The findings underscore the urgent need for continued, expanded genomic surveillance in India. Resistance gene variants detected, the plasmid diversity, and the regional specificities of the isolates all highlight that India’s AMR landscape has features that are not captured by global datasets.
• Understanding how resistance genes are distributed across bacterial species, plasmids, and chromosomes is critical for designing effective infection control strategies. The finding that no pathogen relies solely on chromosomal or plasmid-based resistance means that reducing antibiotic use alone may not eliminate resistance: chromosomal ARGs can persist even after plasmid-borne ones are lost.

More broadly, this work contributes a publicly available genomic dataset of 266 priority pathogen genomes from India to the global scientific community. It is a resource that can support future research, diagnostic development, and policy-making in the fight against AMR.

– Edited by Priyanka/Simran (Research and Development Office).

This blog has been adapted from the original article, available here.