Tkaczyk Eric R, Tkaczyk Alan H, Mauring Koit, Ye Jing Yong, Baker James R, Norris Theodore B
Center for Ultrafast Optical Science, EECS Department, University of Michigan, 2200 Bonisteel Blvd., Ann Arbor, MI 48109-2099, USA.
J Fluoresc. 2009 May;19(3):517-32. doi: 10.1007/s10895-008-0441-z. Epub 2008 Dec 11.
We present a comprehensive study of the selective excitation of two-photon fluorescence from various pairs of dyes and dyes in different conjugation states with tailored pulse shapes found with a genetic algorithm (GA). We investigate a number of biologically important dyes, and include dyes conjugated to trastuzumab (Herceptin(R)) and to a poly(amidoamine) dendrimer. We consider in detail the ability of tailored pulse shaping to discriminate dyes with significant spectral overlap. Our procedure for adaptive pulse shaping includes power-law and chirp-scaling checks to prevent trivial convergences. The GA uses a multiplicative fitness parameter in a graded search method that converges on pulse shapes that not only differentiate two-photon processes, but do so in a high signal regime. We consider the results in terms of not only the absolute maximum ratio of discrimination achieved, but also present the evolutionary course of the GA and compare the improvement to a quantitative measure of the noise level. We also implement a time-domain acousto-optic measurement of two-photon excitation cross-section spectra. The results show that the ability to discriminate dyes is determined almost entirely by their differences in two-photon excitation cross section.
我们展示了一项全面的研究，该研究利用遗传算法（GA）找到的定制脉冲形状，对各种染料对以及处于不同共轭状态的染料的双光子荧光选择性激发进行了研究。我们研究了多种具有生物学重要性的染料，包括与曲妥珠单抗（赫赛汀®）和聚（酰胺胺）树枝状大分子共轭的染料。我们详细考虑了定制脉冲整形区分具有显著光谱重叠的染料的能力。我们的自适应脉冲整形程序包括幂律和啁啾缩放检查，以防止出现平凡收敛。遗传算法在分级搜索方法中使用乘法适应度参数，该方法收敛于不仅能区分双光子过程，而且能在高信号状态下做到这一点的脉冲形状。我们不仅从实现的绝对最大区分率的角度考虑结果，还展示遗传算法的进化过程，并将改进情况与噪声水平的定量测量进行比较。我们还实施了双光子激发截面光谱的时域声光测量。结果表明，区分染料的能力几乎完全由它们在双光子激发截面方面的差异决定。
