Ilija Dukovski

Ilija Dukovski

Auto-generated profile, based on publicly available information

Ilija Dukovski

Bioinformatics Program, Faculty of Computing and Data Sciences, Boston University, Boston, MA, USA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 October 2021 in Nature Protocols
Verified
Journal Article
Closed access
PDF file
IDIlija Dukovski
Ilija Dukovski
DBDjordje Bajić
Djordje Bajić
JCJeremy M. Chacón
Jeremy M. Chacón
MQMichael Quintin
Michael Quintin
JVJean C. C. Vila
Jean C. C. Vila
DSDaniel Segrè
Daniel Segrè

7 October 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
LSLing-Dong Shi
Ling-Dong Shi
MEMaria Florencia Ercoli
Maria Florencia Ercoli
JKJack Kim
Jack Kim
AJArtur Teixeira de Araujo Junior
Artur Teixeira de Araujo Junior
SSSubah Soni
Subah Soni
JBJillian F. Banfield
Jillian F. Banfield
Rice paddies contribute substantially to atmospheric methane (CH 4 ) and these emissions are expected to increase as the need to feed the human population grows. Here, we show that two independent rice genotypes overexpressing genes for PLANT PEPTIDES CONTAINING SULFATED TYROSINE ( PSY ) reduced cumulative CH 4 emissions by 38% (PSY1) and 58% (PSY2) over the growth period compared with controls. Genome-resolved metatranscriptomic data from rhizosphere soils reveal lower ratios of gene activities for CH 4 production versus consumption, decrease in activity of H 2 -producing genes, and increase in bacterial H 2 oxidation pathways in the PSY genotypes. Metabolic modeling using metagenomic and metabolomic data predicts elevated levels of H 2 oxidation and suppressed H 2 production in the PSY rhizosphere. The H 2 -oxidizing bacteria have more genes for utilization of gluconeogenic acids than H 2 -producing counterparts, and their activities were likely stimulated by the observed enrichment of gluconeogenic acids (mostly amino acids) in PSY root exudates. Together these results suggest that decreased CH 4 emission is due to the reduction of H 2 available for hydrogenotrophic methanogenesis. The combination of rice phenotypic characterization, microbiome multi-omic analysis, and metabolic modeling described here provides a powerful strategy to discover the mechanisms by which specific plant genotypes can alter biogeochemical cycles to reduce CH 4 emissions.

12 January 2026 in bioRxiv
Verified
Preprint
Open access
SBSarah Bald
Sarah Bald
JZJing Zhang
Jing Zhang
RNRowan Nelson
Rowan Nelson
DSDillon C. Scott
Dillon C. Scott
IDIlija Dukovski
Ilija Dukovski
DSDaniel Segrè
Daniel Segrè
Synthetic microbial ecology aims at understanding complex microbial dynamics and designing communities with desired properties, often based on mathematical models of individual organisms. However, it is unclear whether simplified models harbor enough detail to predict the composition of synthetic communities, especially in metabolically complex environments. Here, we use longitudinal exometabolite data of monocultures for 15 rhizosphere bacteria to parametrize a consumer-resource model, which we use to predict pairwise co-cultures and higher order communities. The capacity to artificially “switch off” cross-feeding interactions in the model demonstrates their importance in ecosystem structure. Leave-one-out and leave-two-out experiments reveal that pairwise co-cultures do not necessarily capture inter-species interactions within larger communities and broadly highlight the nonlinearity of interactions. Finally, we illustrate that our model can be used to identify new sub-communities of three strains with high likelihood of coexistence. Our results highlight experimental and theoretical challenges to be addressed in future efforts and establish hybrid mechanistic and data-driven metabolic models as a promising framework for predicting and engineering microbial communities.

25 March 2025 in bioRxiv
Verified
Preprint
Open access
PDF file
ZWZoey Werbin
Zoey Werbin
IDIlija Dukovski
Ilija Dukovski
DMDylan Mankel
Dylan Mankel
WAWinston E Anthony
Winston E Anthony
DSDaniel Segrè
Daniel Segrè
MFMarianna Felici
Marianna Felici
Soils harbor diverse microbial communities crucial for ecosystem functioning, but poor genomic representation of many uncultured soil microorganisms limits the utility of existing databases to address some of the most pressing questions in environmental microbiology. To address this, we developed the SoilMicrobeDB, a comprehensive, genome-based reference database to enhance metagenomic classification for soil ecosystems, with a focus on previously underrepresented fungal taxa and uncultured organisms. We evaluated the database using a large soil metagenome dataset, comparing classification rates, analyzing fungal-bacterial ratios against phospholipid fatty acid (PLFA) estimates, and validating lineage abundances with rRNA amplicon sequencing data. Mock community analysis was also conducted to test the precision of community classification and the prevalence of false positives. The SoilMicrobeDB workflow improved metagenomic read classification by over 20% and provided more accurate fungal abundance estimates, particularly for nutrient cycling groups such as ectomycorrhizal fungi. Metagenomic-derived fungal-bacterial ratios were correlated with PLFA and qPCR estimates, and lineage proportions were aligned with relative abundances estimates from rRNA amplicon sequencing. Uncultured taxa represented up to 50% of classifiable soil microbial communities in certain biomes. SoilMicrobeDB offers robust taxonomic and functional profiling of soil communities and provides a scalable and updatable tool for soil microbial ecology research. SoilMicrobeDB is accessible through an interactive platform linking genomes to environmental factors, enabling researchers to explore microbial distributions across soil conditions and potentially leading to new insights into soil ecology and management practices.

14 March 2024 in bioRxiv
Verified
Preprint
Open access
PDF file
IDIlija Dukovski
Ilija Dukovski
LGLauren Golden
Lauren Golden
JZJing Zhang
Jing Zhang
MOMelisa Osborne
Melisa Osborne
DSDaniel Segrè
Daniel Segrè
KKKirill S. Korolev
Kirill S. Korolev
Summary: Microbial colony growth is shaped by the physics of biomass propagation and nutrient diffusion, and by the metabolic reactions that organisms activate as a function of the surrounding environment. While microbial colonies have been explored using minimal models of growth and motility, full integration of biomass propagation and metabolism is still lacking. Here, building upon our framework for Computation of Microbial Ecosystems in Time and Space (COMETS), we combine dynamic flux balance modeling of metabolism with collective biomass propagation and demographic fluctuations to provide nuanced simulations of E. coli colonies. Simulations produced realistic colony morphology, consistent with our experiments. They characterize the transition between smooth and furcated colonies and the decay of genetic diversity. Furthermore, we demonstrate that under certain conditions, biomass can accumulate along “metabolic rings” that are reminiscent of coffee-stain rings, but have a completely different origin. Our approach is a key step towards predictive microbial ecosystems modeling.

4 February 2026 in Journal of The Royal Society Interface
Verified
Journal Article
Open access
ALAlvin Lu
Alvin Lu
LKLiam Kelley
Liam Kelley
IDIlija Dukovski
Ilija Dukovski
DSDaniel Segrè
Daniel Segrè
Viral pathogens, like SARS-CoV-2, hijack the host’s macromolecular production machinery, imposing an energetic burden that is distributed across cellular metabolism. To explore the dynamic metabolic tension between the host’s survival and viral replication, we developed a computational framework that uses genome-scale models to perform dynamic flux balance analysis of human cell metabolism during virus infections. Relative to previous models, our framework addresses the physiology of viral infections of non-proliferating host cells through two new features. First, by incorporating the lipid content of SARS-CoV-2 biomass, we discovered activation of previously overlooked pathways giving rise to new predictions of possible drug targets. Furthermore, we introduce a dynamic model that simulates the partitioning of resources between the virus and the host cell, capturing the extent to which the competition depletes the human cells from essential ATP. By incorporating viral dynamics into our COMETS framework for spatio-temporal modelling of metabolism, we provide a mechanistic, dynamic and generalizable starting point for bridging systems biology modelling with viral pathogenesis. This framework could be extended to broadly incorporate phage dynamics in microbial systems and ecosystems.