Dataset Information

Machine Learning with Optical Phase Signatures for Phenotypic Profiling of Cell Lines.

ABSTRACT: Robust and reproducible profiling of cell lines is essential for phenotypic screening assays. The goals of this study were to determine robust and reproducible optical phase signatures of cell lines for classification with machine learning and to correlate optical phase parameters to motile behavior. Digital holographic microscopy (DHM) reconstructed phase maps of cells from two pairs of cancer and non-cancer cell lines. Seventeen image parameters were extracted from each cell's phase map, used for linear support vector machine learning, and correlated to scratch wound closure and Boyden chamber chemotaxis. The classification accuracy was between 90% and 100% for the six pairwise cell line comparisons. Several phase parameters correlated with wound closure rate and chemotaxis across the four cell lines. The level of cell confluence in culture affected phase parameters in all cell lines tested. Results indicate that optical phase features of cell lines are a robust set of quantitative data of potential utility for phenotypic screening and prediction of motile behavior. © 2019 International Society for Advancement of Cytometry.

SUBMITTER: Lam VK

PROVIDER: S-EPMC7068951 | biostudies-literature | 2019 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Machine Learning with Optical Phase Signatures for Phenotypic Profiling of Cell Lines.

Lam Van K VK Nguyen Thanh T Phan Thuc T Chung Byung-Min BM Nehmetallah George G Raub Christopher B CB

Cytometry. Part A : the journal of the International Society for Analytical Cytology 20190422 7

Robust and reproducible profiling of cell lines is essential for phenotypic screening assays. The goals of this study were to determine robust and reproducible optical phase signatures of cell lines for classification with machine learning and to correlate optical phase parameters to motile behavior. Digital holographic microscopy (DHM) reconstructed phase maps of cells from two pairs of cancer and non-cancer cell lines. Seventeen image parameters were extracted from each cell's phase map, used ...[more]

PMID: 31008570

Similar Datasets

Project description:BackgroundGene expression profiling has revolutionized our ability to molecularly classify primary human tumors and significantly enhanced the development of novel tumor markers and therapies; however, progress in the diagnosis and treatment of melanoma over the past 3 decades has been limited, and there is currently no approved therapy that significantly extends lifespan in patients with advanced disease. Profiling studies of melanoma to date have been inconsistent due to the heterogeneous nature of this malignancy and the limited availability of informative tissue specimens from early stages of disease.Methodology/principle findingsIn order to gain an improved understanding of the molecular basis of melanoma progression, we have compared gene expression profiles from a series of melanoma cell lines representing discrete stages of malignant progression that recapitulate critical characteristics of the primary lesions from which they were derived. Here we describe the unsupervised hierarchical clustering of profiling data from melanoma cell lines and melanocytes. This clustering identifies two distinctive molecular subclasses of melanoma segregating aggressive metastatic tumor cell lines from less-aggressive primary tumor cell lines. Further analysis of expression signatures associated with melanoma progression using functional annotations categorized these transcripts into three classes of genes: 1) Upregulation of activators of cell cycle progression, DNA replication and repair (CDCA2, NCAPH, NCAPG, NCAPG2, PBK, NUSAP1, BIRC5, ESCO2, HELLS, MELK, GINS1, GINS4, RAD54L, TYMS, and DHFR), 2) Loss of genes associated with cellular adhesion and melanocyte differentiation (CDH3, CDH1, c-KIT, PAX3, CITED1/MSG-1, TYR, MELANA, MC1R, and OCA2), 3) Upregulation of genes associated with resistance to apoptosis (BIRC5/survivin). While these broad classes of transcripts have previously been implicated in the progression of melanoma and other malignancies, the specific genes identified within each class of transcripts are novel. In addition, the transcription factor NF-KB was specifically identified as being a potential "master regulator" of melanoma invasion since NF-KB binding sites were identified as consistent consensus sequences within promoters of progression-associated genes.Conclusions/significanceWe conclude that tumor cell lines are a valuable resource for the early identification of gene signatures associated with malignant progression in tumors with significant heterogeneity like melanoma. We further conclude that the development of novel data reduction algorithms for analysis of microarray studies is critical to allow for optimized mining of important, clinically-relevant datasets. It is expected that subsequent validation studies in primary human tissues using such an approach will lead to more rapid translation of such studies to the identification of novel tumor biomarkers and therapeutic targets.

Dataset Information

Machine Learning with Optical Phase Signatures for Phenotypic Profiling of Cell Lines.

Publications

Machine Learning with Optical Phase Signatures for Phenotypic Profiling of Cell Lines.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets