异构多引用对齐:一种单遍方法
Heterogeneous multireference alignment: a single pass approach
摘要:
多参考对齐(MRA)是从许多噪声和循环移位的信号副本中估计信号的问题。在本文中,我们考虑一个扩展称为异构MRA,其中$K$信号必须估计,每个观察来自这些信号之一,我们未知。这是一个简化模型的异质性问题显着出现在冷冻电子显微镜。我们提出了一种算法,估计的$K$信号不估计的位移或类的观察。它只需要对数据进行一次遍历,并且基于在循环移位下不变的低阶矩。给定足够多的测量,可以估计在$K$信号上平均的这些不变特征。然后,我们设计了一个光滑的,非凸的优化问题来计算一组信号,这是一致的估计平均功能。我们发现,在许多情况下,所提出的方法估计的信号集准确,尽管非凸性,和猜想的信号$K$,可以解决作为信号长度$L$的函数的数量是在$\sqrt{L}$的顺序。
英文摘要:
Multireference alignment (MRA) is the problem of estimating a signal from many noisy and cyclically shifted copies of itself. In this paper, we consider an extension called heterogeneous MRA, where $K$ signals must be estimated, and each observation comes from one of those signals, unknown to us. This is a simplified model for the heterogeneity problem notably arising in cryo-electron microscopy. We propose an algorithm which estimates the $K$ signals without estimating either the shifts or the classes of the observations. It requires only one pass over the data and is based on low-order moments that are invariant under cyclic shifts. Given sufficiently many measurements, one can estimate these invariant features averaged over the $K$ signals. We then design a smooth, non-convex optimization problem to compute a set of signals which are consistent with the estimated averaged features. We find that, in many cases, the proposed approach estimates the set of signals accurately despite non-convexity, and conjecture the number of signals $K$ that can be resolved as a function of the signal length $L$ is on the order of $\sqrt{L}$.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。