Deep learning generates custom-made logistic regression models for explaining how breast cancer subtypes are classified.

Differentiating the intrinsic subtypes of breast cancer is crucial for deciding the best treatment strategy. Deep learning can predict the subtypes from genetic information more accurately than conventional statistical methods, but to date, deep learning has not been directly utilized to examine which genes are associated with which subtypes. To clarify the mechanisms embedded in the intrinsic subtypes, we developed an explainable deep learning model called a point-wise linear (PWL) model that generates a custom-made logistic regression for each patient.

Dual inhibition of enhancer of zeste homolog 1/2 overactivates WNT signaling to deplete cancer stem cells in multiple myeloma.

Multiple myeloma (MM) is an incurable hematological malignancy caused by accumulation of abnormal clonal plasma cells. Despite the recent development of novel therapies, relapse of MM eventually occurs as a result of a remaining population of drug-resistant myeloma stem cells. Side population (SP) cells show cancer stem cell-like characteristics in MM; thus, targeting these cells is a promising strategy to completely cure this malignancy.

Japan PGx Data Science Consortium Database: SNPs and HLA genotype data from 2994 Japanese healthy individuals for pharmacogenomics studies.

Japan Pharmacogenomics Data Science Consortium (JPDSC) has assembled a database for conducting pharmacogenomics (PGx) studies in Japanese subjects. The database contains the genotypes of 2.5 million single-nucleotide polymorphisms (SNPs) and 5 human leukocyte antigen loci from 2994 Japanese healthy volunteers, as well as 121 kinds of clinical information, including self-reports, physiological data, hematological data and biochemical data.