DBDNMF: A Dual Branch Deep Neural Matrix Factorization method for drug response prediction

Large-scale information retrieval and correction of noisy pharmacogenomic datasets through residual thresholded deep matrix factorization

Pharmacogenomics studies are attracting an increasing amount of interest from researchers in precision medicine. The advances in high-throughput experiments and multiplexed approaches allow the large-scale quantification of drug sensitivities in molecularly characterized cancer cell lines (CCLs), resulting in a number of open drug sensitivity datasets for drug biomarker discovery. However, a significant inconsistency in drug sensitivity values among these datasets has been noted. Such inconsistency indicates the presence of substantial noise, subsequently hindering downstream analyses. To address the noise in drug sensitivity data, we introduce a robust and scalable deep learning framework, Residual Thresholded Deep Matrix Factorization (RT-DMF). This method takes a single drug sensitivity data matrix as its sole input and outputs a corrected and imputed matrix. Deep matrix factorization (DMF) excels at uncovering subtle patterns, due to its minimal reliance on data structure assumptions. This attribute significantly boosts DMF's ability to identify complex hidden patterns among nuisance effects in the data, thereby facilitating the detection of signals that are therapeutically relevant. Furthermore, RT-DMF incorporates an iterative residual thresholding procedure, which plays a crucial role in retaining signals more likely to hold therapeutic importance. Validation using simulated datasets and real pharmacogenomics datasets demonstrates the effectiveness of our approach in correcting noise and imputing missing data in drug sensitivity datasets (open-source package available at https://github.com/tomwhoooo/rtdmf).

Differential bone morphology and hypoxia activity in skeletal metastases of ER+ and ER- breast cancer

Bone metastases occur in the majority of advanced breast cancer patients, and the most common complications are osteolytic bone metastases. However, due to the limitations of models and methodologies for bone metastasis studies, the intricacies of bone metastasis have not been fully acknowledged and revealed in prior work. Our earlier study indicated that certain breast cancer cells undergo a pre-osteolytic stage before the establishment of overt metastatic lesions. Here, we further identify a differential bone morphology between ER (estrogen receptor)+ and ER- breast cancer. Specifically, we observe a more pronounced osteolytic phenotype in the bone metastatic lesions of ER- cells investigated, linked to elevated hypoxia signaling that stimulates the secretion of osteolytic inducers from cancer cells. In the in vivo mouse model, the application of the hypoxia-inducible factor (HIF) inhibitor 2-methoxyestradiol demonstrates a promising efficacy in suppressing tumor growth and osteoclast differentiation in the bone lesions established by bone-tropic subpopulation of ER- MDA-MB-231 cell. Overall, our findings explore the complexity of phenotype and morphology in bone metastatic lesions of ER+ and ER- breast cancer, which offers a compelling rationale for leveraging HIF inhibitors to the treatment targeting skeletal complications of breast cancer bone metastases, especially for ER- tumors.