The Arabidopsis thaliana Kinesin-5 AtKRP125b Is a Processive, Microtubule-Sliding Motor Protein with Putative Plant-Specific Functions.

The formation and maintenance of the mitotic spindle during cell division requires several microtubule-interacting motor proteins. Members of the kinesin-5 family play an essential role in the bipolar organization of the spindle. These highly conserved, homotetrameric proteins cross-link anti-parallel microtubules and slide them apart to elongate the spindle during the equal separation of chromosomes. Whereas vertebrate kinesin-5 proteins are well studied, knowledge about the biochemical properties and the function of plant kinesin-5 proteins is still limited. Here, we characterized the properties of AtKRP125b, one of four kinesin-5 proteins in Arabidopsis thaliana. In in vitro motility assays, AtKRP125b displayed the archetypal characteristics of a kinesin-5 protein, a low velocity of about 20 nm·s-1, and a plus end-directed, processive movement. Moreover, AtKRP125b was able to cross-link microtubules and to slide them apart, as required for developing and maintaining the mitotic spindle. In line with such a function, GFP-AtKRP125b fusion proteins were predominantly detected in the nucleus when expressed in Arabidopsis thaliana leaf protoplasts or Nicotiana benthamiana epidermis cells and analyzed by confocal microscopy. However, we also detected GFP signals in the cytoplasm, suggesting additional functions. By generating and analyzing AtKRP125b promoter-reporter lines, we showed that the AtKRP125b promoter was active in the vascular tissue of roots, lateral roots, cotyledons, and true leaves. Remarkably, we could not detect promoter activity in meristematic tissues. Taken together, our biochemical data support a role of AtKRP125b in mitosis, but it may also have additional functions outside the nucleus and during interphase.

Regular expressions for decoding of neural network outputs.

This article proposes a convenient tool for decoding the output of neural networks trained by Connectionist Temporal Classification (CTC) for handwritten text recognition. We use regular expressions to describe the complex structures expected in the writing. The corresponding finite automata are employed to build a decoder. We analyze theoretically which calculations are relevant and which can be avoided. A great speed-up results from an approximation. We conclude that the approximation most likely fails if the regular expression does not match the ground truth which is not harmful for many applications since the low probability will be even underestimated. The proposed decoder is very efficient compared to other decoding methods. The variety of applications reaches from information retrieval to full text recognition. We refer to applications where we integrated the proposed decoder successfully.

Design strategies for weight matrices of echo state networks.

This article develops approaches to generate dynamical reservoirs of echo state networks with desired properties reducing the amount of randomness. It is possible to create weight matrices with a predefined singular value spectrum. The procedure guarantees stability (echo state property). We prove the minimization of the impact of noise on the training process. The resulting reservoir types are strongly related to reservoirs already known in the literature. Our experiments show that well-chosen input weights can improve performance.