Embedded Files
Systematic characterization and development of mammalian cell-based bioprocesses
Systematic characterization and development of mammalian cell-based bioprocesses
We have earlier built the first-ever genome-scale model of Chinese hamster ovary (CHO) cells (Hefzi et al. 2017), a biopharmaceutical workhorse cell line, and have further refined it by accounting for enzyme capacity (Yeo et al. 2020). We have also demonstrated its applicability to characterize industrial processes (Hong et al. 2020). We now work towards using these models to design mammalian cell-based processes to improve the yield and quality. Particularly, we use the model-based approach to design and optimize cell culture media to rationally improve cell growth and titer (Yeo et al. 2022). We also develop new mechanistic models for various cellular processes such as N-glycosylation and other post-translational modifications as well as viral replication/infection. We further develop frameworks to hybridize such individual cellular models using AI/ML (Lakshmanan et al. 2024). This will enable the prediction of quality profiles of the manufactured drug in addition to titer, and thus enabling its control through media and process designs. Our group is also interested in the development of whole bioprocess models benefit in assessing scale-up/scale-down, techno economics and sustainability.
Apart from biopharmaceutical manufacturing, we also apply similar concepts to optimize the cultivated meat and cell therapy products' cell culture media as well as the associated bioprocess.
Data-driven knowledge-based design of mammalian cell lines for biomanufacturing
Data-driven knowledge-based design of mammalian cell lines for biomanufacturing
We have earlier developed multi-omics data driven frameworks for rational cell line development and engineering and have successfully applied this framework to decipher 1) the genomic changes when a CHO cell becomes antibody producer (Yusufi et al. 2017) and 2) genomic and phenotypic traits of various CHO parental cell lines (Lakshmanan et al. 2019). We now work on frameworks to rationalize mammalian cell line development by identifying stable genomic integration sites to further the transgene expression. Moreover, we also mine various mammalian epigenome data such as ATAC-seq and DNAse-seq to identify the active endogenous enhancers and understand the gene expression machinery, and thereby improving the vector reconstruction and development.
Systematic elucidation and design of probiotics
Systematic elucidation and design of probiotics
In this workflow, we plan to expand our newly developed framework for probiotic evaluation by evaluating unconventional bacteria found in human gut for probiotic applications (Koduru et al. 2022) . We are particularly interested in evaluating multiple species from following genus: Akkermansia, Bacteroides, Faecalibacterium and Roseburia. We shall also expand by exploring the interactions resulting from the multiple gut microbial ligand-host receptor combinations due to the small molecules produced by these probiotic species. We may use the available models for other gut microbial species as well as the host organisms such as human and mouse.
Novel constraint-based modeling and multi-omics integration algorithm/tool development
Novel constraint-based modeling and multi-omics integration algorithm/tool development
We develop new methods to integrate multi-omics data and constraint-based modeling. We particularly focus on integrating single cell data with models to study metabolic variations at single cell level in diseases.
Single-cell techniques have become key tools in probing cellular heterogeneity. Leveraging on single cell data to simulate genome-scale metabolic models (GEMs) is a crucial step in relating gene expression to metabolic activities at different cell states. However, the systematic noise and data sparsity in single cell RNAseq (scRNAseq) data complicates such integration. We develop a new method that combines probabilistic imputation to compensate the dropouts and gene expression-weighted context-specific GEM reconstruction in a single step. The ensemble of GEMs generated enables an unbiased estimation of metabolic fluxes corresponding to single-cell states.
We also develop methods to identify gene targets to be targeted for differentiation steps by mining and evaluating the scRNAseq data obtained from the gestation process or other in vitro differentiation processes. The major application areas of such algorithms include cultivated meat and cell therapy applications.
References
References
Hefzi et al. (2016) “A consensus genome-scale reconstruction of CHO cell metabolism for improved biotherapeutic protein production”, Cell Systems, 3(5):434-443.
Hong et al. (2020) “In silico model-based characterization of metabolic response to harsh sparging stress in fed-batch CHO cell cultures”, Journal of Biotechnology, 308:10-20.
Koduru et al. (2022) “Systematic evaluation of genome-wide metabolic landscapes in lactic acid bacteria reveals diet-induced and strain-specific probiotic idiosyncrasies”, Cell Reports, 41(10): 111735.
Lakshmanan et al. (2019) “Multi-omics profiling of CHO parental hosts reveals cell line-specific variations in bioprocessing traits”, Biotechnology & Bioengineering, 116(9):2117-2129.
Lakshmanan et al. (2024) " Antibody glycan quality predicted from CHO cell culture media markers and machine learning", Computational and Structural Biotechnology Journal, 23:2497-2506.
Yeo et al. (2020) “Enzyme capacity–based genome-scale modelling of CHO cells”, Metabolic Engineering, 60:138–147.
Yeo et al. (2022) “Combined multivariate statistical and flux balance analyses uncover media bottlenecks to the growth and productivity of CHO cell cultures”, Biotechnology & Bioengineering, 119(7):1740-1754.
Yusufi et al. (2017) “Mammalian systems biotechnology reveals global cellular adaptations in a recombinant CHO cell line”, Cell Systems, 4(5):530-542.
Page updated
Google Sites
Report abuse