The focus of the group is on developing models to gain insights into problems arising in biology and medicine. In the last few years, the group has focused on the following areas: cancer, neurodegeneration, and cultured meat.
In cancer, we focus on modeling the biophysical aspects of cell migration. During cancer metastasis, cancer cells migrate and spread to a distant part of the body to initiate the formation of a new tumor. We develop novel assays to measure the biophysical properties of cell migration and elucidate the signaling and gene regulatory networks that govern these properties, especially the role of the cytoskeleton. From these studies, we hope to identify novel therapeutic targets to perturb cell migration. In particular, it has been shown that, during breast cancer progression, the deposition of extracellular matrix (ECM) proteins such as collagen increases, resulting in higher matrix rigidity. However, it Is still not known how ECM rigidity can influence tumor mass growth, as cancer cells have to proliferate and migrate against mechanical stresses imposed by the ECM. We are investigating the effects of matrix rigidity on the growth rate of breast tumor spheroids, by embedding multicellular spheroids of MCF7, a breast cancer cell line, in collagen gels of various rigidity, and then measuring MCF7 spheroid growth rate by quantifying images obtained from confocal microscopy (Fig. 1). We are also studying the mechanisms for this rigidity dependence. For example, we found that tumor spheroid growth is determined by actomyosin tension which may alter cell rounding pressure during cell division.
While the production of cultured meat may seem to be a dramatically different problem than cancer, the underlying biophysics are actually remarkably similar. Cultured meat refers to meat grown by culturing animal cells directly. It is becoming a promising area of research that offers potential to address issues such as environmental pollution, disease outbreaks, food security and animal welfare. Individual muscle cells or myoblasts have to migrate and fuse to form myotubes, which in turn come together to form myofibers. The ability to direct the formation of these hierarchical structures is an important step in cultured meat problem. We seek to develop methods to understand and optimize this process, by developing biophysical models to understand the mechanisms of nuclei centration and spreading during myotube formation. We seek to optimize the topography of the microenvironment to maximize myotube formation for cultured meat production.
Figure. 1. Confocal images of breast tumor spheroids embedded in increasing concentrations of collagen in a 3D matrix. Nuclei are stained in blue (DAPI), and F-actin in red. Quantification of the volume of spheroids and number of nuclei embedded in different collagen concentrations.
In addition, cultured meat production requires inexpensive serum-free media that supports animal cell growth. We are also developing methods based on genetic algorithms that allow fast iterative optimization of media component concentration. We have demonstrated ~30% improvement in cell proliferation using our optimization algorithm (Fig. 2).
Finally, for neurodegeneration, we are interested in discovering biomarkers for early onset of neurodegenerative diseases such as Alzheimer’s Disease (AD). The definitive diagnosis of AD without the need for neuropathological confirmation remains a challenge in AD research today, despite efforts to uncover the molecular and biological underpinnings of the disease process. Furthermore, the potential for therapeutic intervention is limited upon the onset of symptoms. Therefore, there is strong motivation to studying and treating the AD precursor mild cognitive impairment (MCI), the prodromal stage of AD, instead. We apply machine learning classification to transcriptomic data of MCI, AD, and cognitively normal (CN) control patients, and identified differentially expressed genes that serve as biomarkers for the characterization and classification of subjects into MCI or AD groups. Predictive models employing these biomarker genes exhibited good classification performances for CN, MCI, and AD. Furthermore, we found that the PI3K-Akt, IL-17, JAKSTAT, TNF, and Ras signaling pathways were also enriched in these biomarker genes, indicating their diagnostic potential and pathophysiological roles in MCI and AD. We believe that these findings could aid in the recognition of MCI and AD risk in clinical settings, and allow for the tracking of disease progression over time in individuals as part of a therapeutic Cellular Image Informatics Division (CIID) BIOINFORMATICS INSTITUTE 2023 29 1. Jeremy Joon Ho Goh, Corinna Jie Hui Goh, Qian Wei Lim, Songjing Zhang, Cheng-Gee Koh, K.-H. Chiam, Transcriptomics indicate nuclear division and cell adhesion not recapitulated in MCF7 and MCF10A compared to luminal A breast tumours, Scientific Reports 12, 20902 (2022) 2. Sher Li Oh, Meikun Zhou, Eunice E. W. Chin, Gautami Amarnath, Chee Hoe Cheah, Kok Pin Ng, Nagaendran Kandiah, Eyleen L Goh, K.-H. Chiam, Alzheimer’s Disease Blood Biomarkers Associated with Neuroinflammation as Therapeutic Targets for Early Personalized Intervention, Frontiers in Digital Health 4, 875895 (2022) 3. Lor Huai Chong, Terry Ching, Hui Jia Farm, Gianluca Grenci, K.-H. Chiam, Yi-Chin Toh, Integration of a microfluidic multicellular coculture array with machine learning analysis to predict adverse cutaneous drug reactions, Lab on a Chip 22, 1890 (2022) SELECTED PUBLICATIONS 4. Shingo Tsukamoto, K.-H. Chiam, Takumi Asakawa, Kaoru Sawasaki, Naoyuki Takesue, Naoya Sakamoto, Compressive forces driven by lateral actin fibers are a key to the nuclear deformation under uniaxial cell-substrate stretching, Biochemical and Biophysical Research Communications 597, 37 (2022) 5. Ai Kia Yip, Songjing Zhang, Lor Huai Chong, Elsie Cheruba, Jessie Yong Xing Woon, Theng Xuan Chua, Corinna Jie Hui Goh, Haibo Yang, Chor Yong Tay, Cheng-Gee Koh, K.-H. Chiam, Zyxin Is Involved in Fibroblast Rigidity Sensing and Durotaxis, Frontiers in Cell and Developmental Biology 9, 3264 (2021) 6. Nicole Zi-Jia Khong, Yukai Zeng, Soak-Kuan Lai, Cheng-Gee Koh, Zhao- Xun Liang, K.-H. Chiam*, Hoi-Yeung Li*, Dynamic swimming patterns of Pseudomonas aeruginosa near a vertical wall during initial attachment stages of biofilm formation, Scientific Reports 11, 1952 (2021) Chiam Keng Hwee is a theorist working at the interface of physics and biology, collaborating very closely with experimental groups in developing theories and models for a variety of problems in mechanobiology and biological physics, systems biology, and biological fluid mechanics. He received his Ph.D. in physics from the California Institute of Technology in 2003 and his B.S.E. in physics from the University of Michigan in 1997. PRINCIPAL INVESTIGATOR’S BIOGRAPHY approach, and provide possible personalized drug targets for early intervention of MCI and AD.
Figure 2. Workflow for cell culture media optimization.
We also use computer vision and machine learning methods to analyze behaviors of animals for early detection of neurodegeneration. Automatic tracking and analysis of behaviors from videos can enable higher throughput studies and more quantitative understanding of how neurological diseases affects animals’ individual and social behaviors. This can potentially be used in applications such as treatment testing in animal disease models, and development of diagnosis systems for neurological diseases from videos. Behavior classification from pose extracted from videos is becoming popular in animal behavior research. We develop self-supervised learning approach on pose estimation data to enable more learning from unlabeled data that improves subsequent supervised learning performance with limited labeled data (Fig. 3).
Figure 3. Workflow for self-supervised learning for animal behavioral classification.
Chiam Keng Hwee is a theorist working at the interface of physics and biology, collaborating very closely with experimental groups in developing theories and models for a variety of problems in mechanobiology and biological physics, systems biology, and biological fluid mechanics. He received his Ph.D. in physics from the California Institute of Technology in 2003 and his B.S.E. in physics from the University of Michigan in 1997.
Keng Hwee's group uses a combination of biophysical and bioinformatics tools to study cell migration. Cell migration is a critical process in every living organism, central to, for example, the morphogenesis of embryos, formation of tissues and organs, wound repair, as well as in less welcoming scenarios such as cancer metastasis. Some of their current projects include the study of amoeboid modes of cancer cell invasion, collective modes of epithelial cell sheet migration, and swimming and swarming modes of bacterial cell motility.
He hopes that their approach will enable them to identify potential targets to perturb cell migration, which can eventually be translated into drugs to stop cancer cell invasion, promote wound and skin healing, or stop the aggregation of bacterial cells into biofilms.
From groundbreaking discoveries to cutting-edge research, our researchers are empowering the next generation of female science, technology, engineering and mathematics (STEM) leaders. Get inspired by our #WomeninSTEM