Confinement loss prediction in diverse anti-resonant fibers through neural networks.

In this work, genetic algorithm (GA) is employed to optimize convolutional neural networks (CNNs) for predicting the confinement loss (CL) in anti-resonant fibers (ARFs), achieving a prediction accuracy of CL magnitude reached 90.6%, which, to the best of our knowledge, represents the highest accuracy to date and marks the first instance of using a single model to predict CL across diverse ARF structures. Different from the previous definition of ARF structures with parameter groups, we use anchor points to describe these structures, thus eliminating the differences in expression among them. This improvement allows the model to gain insight into the specific structural characteristics, thereby enhancing its generalization capabilities. Furthermore, we demonstrate a particle swarm optimization algorithm (PSO), driven by our model, for the design of ARFs, validating the model's robust predictive accuracy and versatility. Compared with the calculation of CL by finite element method (FEM), this model significantly reduces the cost time, and provides a speed-up method in fiber design driven by numerical calculation.

Antiresonant fiber structures based on swarm intelligence design.

In this work, we obtained a new, to the best of our knowledge, structure of anti-resonant fiber (ARF) by an adaptive particle swarm optimization (PSO) algorithm. Different from the prior method of stacking elemental parts and optimizing parameters through experience or algorithm, we decompose some classic structures into points and optimize the positions of these points through swarm intelligence. The fiber structure is reconstructed by interpolation, and some new structures with low confinement loss (CL) and high higher order mode extinction ratio (HOMER) are obtained. These novel ARFs exhibit similar structural characteristics, and are named as "the bulb-shaped ARFs". Among these structures, the minimum achieved CL is 2.21 × 10-5dB/m at 1300 nm and the maximum achieved HOMER exceeds 14,000. This work provides a method with high degree of freedom in the design of non-uniform cross-section waveguides and helps to discover new fiber structures.

A heading date QTL, qHD7.2, from wild rice (Oryza rufipogon) delays flowering and shortens panicle length under long-day conditions.

Heading date (HD) and panicle length (PL) are important traits that affect rice breeding and are controlled by pleiotropic genes. Some alleles associated with HD and PL from wild relatives might differ from those in cultivated rice. In this study, a main effect HD quantitative trait locus from wild rice, qHD7.2, was identified using a chromosomal segment substitution line (CSSL) population. First, qHD7.2 was determined to be located near RM172 on chromosome 7 based on association analysis of phenotype data from six environments and 181 polymorphic molecular markers. CSSL39, which has the latest flowering of all CSSLs and carries qHD7.2, was selected for further study, and qHD7.2 was narrowed to a 101.1-kb interval using a CSSL39/9311 F2 population. An OsPRR37-homologous gene was found within this region. The wild type allele delayed flowering and shortened PL under long-day conditions. The HD7.2, which was identified as a candidate gene for qHD7.2, transcript level was substantially higher than that in 9311. Our data showed that HD7.2 is likely a novel OsPRR37 allele. Sequence analysis revealed that OsPRR37 in cultivated rice had multiple origins, and natural variation in the coding domain sequence and promoter region contribute to flowering time diversity in cultivated rice.