Artificial Intelligence-Driven Morphology-Based Enrichment of Malignant Cells from Body Fluid.

Cell morphology is a fundamental feature used to evaluate patient specimens in pathologic analysis. However, traditional cytopathology analysis of patient effusion samples is limited by low tumor cell abundance coupled with the high background of nonmalignant cells, restricting the ability of downstream molecular and functional analyses to identify actionable therapeutic targets. We applied the Deepcell platform that combines microfluidic sorting, brightfield imaging, and real-time deep learning interpretations based on multidimensional morphology to enrich carcinoma cells from malignant effusions without cell staining or labels. Carcinoma cell enrichment was validated with whole genome sequencing and targeted mutation analysis, which showed a higher sensitivity for detection of tumor fractions and critical somatic variant mutations that were initially at low levels or undetectable in presort patient samples. Our study demonstrates the feasibility and added value of supplementing traditional morphology-based cytology with deep learning, multidimensional morphology analysis, and microfluidic sorting.

Large-scale, high-throughput validation of short hairpin RNA sequences for RNA interference.

A method for high-throughput cloning and analysis of short hairpin RNAs (shRNAs) is described. Using this approach, 464 shRNAs against 116 different genes were screened for knockdown efficacy, enabling rapid identification of effective shRNAs against 74 genes. Statistical analysis of the effects of various criteria on the activity of the shRNAs confirmed that some of the rules thought to govern small interfering RNA (siRNA) activity also apply to shRNAs. These include moderate GC content, absence of internal hairpins, and asymmetric thermal stability. However, the authors did not find strong support for position specific rules. In addition, analysis of the data suggests that not all genes are equally susceptible to RNA interference (RNAi).

Generation of bioreagents for protein chips.

Protein microarrays have the potential to dramatically increase the throughput of proteomic analysis. Protein expression profiling chips with distinct spots of immobilized protein capture agents will allow the simultaneous measurement of hundreds to thousands of proteins from one sample. In contrast to DNA chips, for which the capture probes are easily designed and synthesized, the development of content for protein biochips is a long and laborious process. Careful consideration must be given to the specificities desired, the format of the assay, and the requirements of the capture agents, as well as to process optimization to minimize development time and cost. Monoclonal antibodies have been the prime choice as protein capture agents for the majority of protein chips developed to date. New technologies for the production of protein capture agents are more amenable to automation than traditional monoclonal antibody production and therefore carry the promise for industrialization.