Lipid droplets are dynamic organelles that play important cellular roles. They are composed of a phospholipid membrane and a core of triglycerides and sterol esters. Fatty acids have important roles in phospholipid membrane formation, signaling, and synthesis of triglycerides as energy storage. Better non-invasive tools for profiling and measuring cellular lipids are needed. Here we demonstrate the potential of Raman spectroscopy to determine with high accuracy the composition changes of the fatty acids and cholesterol found in the lipid droplets of prostate cancer cells treated with various fatty acids. The methodology uses a modified least squares fitting (LSF) routine that uses highly discriminatory wavenumbers between the fatty acids present in the sample using a support vector machine algorithm. Using this new LSF routine, Raman micro-spectroscopy can become a better non-invasive tool for profiling and measuring fatty acids and cholesterol for cancer biology.

中文翻译：

---

使用拉曼和改进的支持向量机算法进行脂质分析

脂滴是具有重要细胞作用的动态细胞器。它们由磷脂膜和甘油三酯和甾醇酯的核心组成。脂肪酸在磷脂膜的形成、信号传导和甘油三酯的合成中作为能量储存具有重要作用。需要更好的非侵入性工具来分析和测量细胞脂质。在这里，我们展示了拉曼光谱在高精度确定用各种脂肪酸处理的前列腺癌细胞的脂滴中发现的脂肪酸和胆固醇的组成变化的潜力。该方法使用改进的最小二乘拟合 (LSF) 例程，该例程使用支持向量机算法在样本中存在的脂肪酸之间使用高度区分的波数。使用这个新的 LSF 例程，