Metagenomic Sequencing Reveals a Hidden World of Bacteriophages in the Ganga River

The Ganga River, which stretches 2,525 kilometers from the Gangotri glacier in the Himalayas across Uttarakhand, Uttar Pradesh, Bihar and West Bengal, sustains the most populous river basin on Earth. It is also a river under severe ecological stress, receiving continuous discharges of untreated industrial and household effluents that have raised its chemical and microbial loads over recent decades. Yet beneath its polluted surface lies a viral universe that has remained largely unmapped. A new metagenomic study, published in International Microbiology, has now sequenced the viral DNA and RNA present in the river’s water at Haridwar, offering one of the first detailed portraits of the bacteriophage communities that inhabit this iconic freshwater ecosystem.

The research team collected water samples from three sites along the Haridwar stretch of the river, all chosen for their high levels of human activity: the Daksheshwar Mahadev temple, the famous bathing ghat of Har ki Pauri, and Chandi Devi Ghat along the Eastern Ganga canal road. Samples from all three locations were gathered on the same day, transported at 4 degrees Celsius in sterile centrifuge tubes, and pooled in equal proportions before being stored at minus 20 degrees Celsius. From this pooled sample, the researchers extracted both metagenomic DNA and RNA using commercial kits, verifying quality by gel electrophoresis and spectrophotometry before constructing sequencing libraries.

The laboratory workflow followed established metagenomic protocols. DNA and RNA-derived cDNA were fragmented into roughly 600 base-pair segments using the KAPA HyperPlus kit, end-repaired, A-tailed, and ligated to sequencing adapters. After library amplification with Illumina primers and purification with Ampure beads, the libraries were quantified with a Qubit high-sensitivity assay and sequenced on an Illumina HiSeq 4000 platform. The run produced 22.4 million paired-end sequences of 151 base pairs each, with a GC content of 59 percent and low duplication rates of 4.9 percent for forward reads and 2.5 percent for reverse reads, indicating a robust and technically consistent sequencing depth suitable for downstream analysis.

Bioinformatic processing began with quality assessment using FastQC, followed by trimming and filtering with fastp and Trim Galore. Taxonomic classification was then performed with Kraken 2, a k-mer-based classifier that matches short sequence fragments against a reference database of complete bacterial, archaeal and viral genomes from NCBI RefSeq. Low-abundance taxa represented by fewer than ten reads or less than 0.01 percent of classified reads were excluded to remove likely false positives. Phylogenetic relationships among the identified phages were reconstructed using ETE 3 for visualization and VICTOR, the genome-based virus classification service of the Leibniz Institute DSMZ. The raw sequence data have been deposited in the NCBI Sequence Read Archive under BioProject accession PRJNA1207791.

The results revealed a strikingly diverse viral community. In total, the sample contained 6,543 distinct viral species, from which 549,515 sequence reads were obtained. Seventeen viral phyla were identified, dominated overwhelmingly by Uroviricota, the phylum that encompasses tailed bacteriophages, with 3,575 hits. Negarnaviricota and Pisuviricota followed with 474 and 460 reads respectively, while Artverviricota and Kitrinoviricota contributed 361 and 338. At the class level, Caudoviricetes, the tailed phages, stood out dramatically with 3,575 bacteriophages, far ahead of Revtraviricetes at 360 and Pisoniviricetes at 287. Among classified orders, Crassvirales led with 867 hits and Kirjokansivirales followed with 795, but the single largest category was the unclassified group, containing 1,488 reads, a clear sign that a substantial fraction of the river’s virome remains scientifically undescribed.

At the family level, one group dominated the dataset to a remarkable degree. Casjensviridae accounted for 516,163 generated sequences, dwarfing every other phage family; Rountreeviridae followed with 10,094 sequences, then Peduoviridae with 5,734 and Autographiviridae with 2,614. Kyanoviridae, Herelleviridae, Schitoviridae, Demerecviridae, Mesyanzhinovviridae and Crevaviridae completed the top ten. Among genera, the unclassified category again led with 300,671 sequences, followed by Chivirus with 205,535 and Copernicusvirus with 8,607. At the species level, the Salmonella phage Chivirus iEPS5 was the most abundant organism in the entire sample with 300,245 reads, followed by the Vibrio phage Valbvirus ValB1MD2 at 124,213 reads and the Staphylococcus phage Andhravirus andhra at 68,758. Abundance fell steeply beyond these top few, indicating a strongly skewed phage population structure.

Many of the dominant phages have well-characterized host ranges documented in earlier studies, and their presence in the Ganga carries ecological significance. Chivirus iEPS5 is a flagellatropic Salmonella phage that depends on the counterclockwise-rotating flagellar filament of Salmonella enterica serovar Typhimurium for adsorption and injects its DNA through the filament during infection. ValB1MD2 kills Vibrio alginolyticus, an emerging foodborne pathogen linked to contaminated seafood. Andhravirus andhra, originally classified among the podoviruses, infects Staphylococcus epidermidis, and its proteins Andhra_gp10 and Andhra_gp14 have been shown to inhibit bacterial growth and disintegrate staphylococcal cell walls. Rosenblumvirus CSA13, another Staphylococcus phage, can prevent biofilm formation across multiple environments, marking it as a candidate biocontrol agent.

The phylogenetic analysis added further texture to the picture. Five Pandoravirus species appeared among the 25 most prevalent viral species, making it the most species-rich genus in that set. Chivirus iEPS5, Zhonglingvirus SAP012, Andhravirus andhra and Rosenblumvirus CSA13 all belong to the class Caudoviricetes, with the first two sharing the family Casjensviridae and thus a closer taxonomic relationship. Other caudoviricetes including Elvirus EL, Efquatrovirus LY0322, which infects the drug-resistant bacterium Enterococcus faecalis, the Agrobacterium phage Atu_ph07 and the Pseudomonas phage Psa21 formed related clusters. Psa21 has previously been recognized as a potential biocontrol agent against bacterial canker of kiwifruit, while Atu_ph07 specifically infects Agrobacterium tumefaciens, the plant pathogen responsible for crown gall disease.

The study’s context stretches back more than a century. In 1896, the bacteriologist Ernest Hanbury Hankin reported antibacterial activity in the waters of the Ganga and Yamuna, noting that the activity was filterable and heat-labile and limited the spread of cholera, an observation made decades before the formal discovery of bacteriophages. Today, with an estimated 10 to the power of 31 viruses on Earth, phages are understood as key regulators of microbial populations and environmental processes. The authors note that the Ganga’s phages, each with particular host bacterial species, appear to reduce the potency and lethality of their pathogenic hosts, and that phage genes linked to bacterial virulence and antibiotic resistance underscore the ecological role of phages in shaping microbial dynamics and bacterial evolution in the river.

The findings arrive at a moment when phage therapy is regaining momentum as a response to multidrug-resistant bacteria, since lytic phages can selectively destroy infected bacterial cells and multiply within their hosts, advantages that become critical as antibiotic resistance genes spread through aquatic environments. The authors caution that their approach has limitations: Kraken 2 relies on exact k-mer matches and can therefore miss divergent or erroneous sequences, read-level classification offers limited resolution at the species level in complex metagenomes, and read counts as proxies for abundance are biased toward larger viral genomes that generate more mappable fragments. Even so, the work provides a genomic baseline for the phageome of the upper Ganga, a system never analyzed in this way before, and points toward phage-driven strategies for environmental monitoring, pollution management and the eventual development of bacteriophage-based therapeutic interventions rooted in the river’s own viral diversity.

Subject of Research: Bacteriophage diversity in the Ganga River ecosystem revealed by metagenomic sequencing

Article Title: Analysis of bacteriophage diversity in Ganga river ecosystem: insights from metagenome-based sequencing

Article References: Katara, A., Chand, S., Chaudhry, V., Vishwakarma, S., Dubey, R. C., Chandra, H., Shilbayeh, S. A. R., Khan, S.-U.-D., Vohra, S., Poddar, N. K., & Khan, S. (2026). Analysis of bacteriophage diversity in Ganga river ecosystem: insights from metagenome-based sequencing. International Microbiology. https://doi.org/10.1007/s10123-026-00901-5

Image Credits: AI Generated

DOI: 10.1007/s10123-026-00901-5

Keywords: bacteriophages, metagenomics, Ganga River, virome, Caudoviricetes, Casjensviridae, water microbiology, phage therapy, antibiotic resistance, viral ecology, freshwater ecosystem, Haridwar