Spectroscopy Assisted Laser Microdissection

摘要 对肿瘤的分子理解在很大程度上依赖于从上皮细胞制备的适当样品， 组织中然而，上皮细胞通常被其他细胞类型包围，并提取纯的细胞群。 这些细胞对于正确的生物样本制备和分子测定的准确性至关重要。激光 显微切割（LM）由于其在提取中的高度空间特异性，在这方面做出了巨大贡献 明确的细胞群和易用性。LM在提高生化分析精度的同时， 仍然存在一些缺点。染色和人工监督的必要性限制了通量、分子产量 和样品的纯度。对提取的细胞群的纯度几乎没有明确的控制或信心 同时难以从同一样品中提取多个细胞。结合了 显微镜和光谱学的分子灵敏度，红外（IR）光谱成像被用来 使用人工智能算法在复杂组织中自动组织病理学识别 光谱数据这个项目将展示一个完全自动化的仪器耦合LM与IR 显微镜被称为光谱辅助激光显微切割（SLaM），开发的原型将是 使用最先进的红外成像系统和商用LCM在精度、速度和类型方面进行验证 忠诚最后，该方法将被应用于从同一前列腺样本中提取不同类型的细胞， 证明了从相同组织多重LM（muxLM）的能力。该项目直接针对 需要减少LM的时间和劳动密集性。SLaM可以最大限度地提高 用于下游分析的生物样品通过自动化、高通量和精确度， 细胞的全面采集，没有用户疲劳或错误，从而提供更高完整性的样品 和癌症分子分析的质量。

Abstract Molecular understanding of tumors relies greatly on appropriate samples to be prepared from epithelial cells in tissues. Epithelial cells, however, are often surrounded by other cell types and extracting pure populations of these cells is crucial for correct biospecimen preparation and resulting accuracy of molecular assays. Laser microdissection (LM) has contributed immensely in this effort due to its high spatial specificity in the extraction of defined cell populations and ease of use. While LM has enhanced the precision of biochemical analysis, several drawbacks remain. The necessity of staining and human supervision limits throughput, molecular yield and purity of samples. There is little explicit control or confidence in the purity of extracted cell populations while it is difficult to extract multiple cells from the same sample. Combining the morphologic specificity of microscopy and molecular sensitivity of spectroscopy, infrared (IR) spectroscopic imaging been employed to automate histopathologic recognition in complex tissues using artificial intelligence algorithms applied of spectral data. This project will demonstrate a completely automated instrument by coupling LM with IR microscopy. Termed spectroscopy-assisted laser microdissection (SLaM), the developed prototype will be validated using state of the art IR imaging systems and commercial LCM in terms of accuracy, speed and type fidelity. Last, the approach will be applied to extract cells of different types from the same prostate sample to demonstrate the capability to multiplex LM (muxLM) from the same tissue. The project directly addresses the need to reduce the time- and labor-intensive nature of LM. SLaM can maximize the quality and utility of biological samples used for downstream analyses by automation, high throughput and precision while enabling a comprehensive acquisition of cells without user fatigue or error, thereby providing a sample of higher integrity and quality for cancer molecular analysis.