Gene essentiality is typically determined by assessing the viability of the corresponding mutant cells, but this definition fails to account for the ability of cells to adaptively evolve to genetic perturbations. Here, we performed a stringent screen to assess the degree to which Saccharomyces cerevisiae cells can survive the deletion of ~1,000 individual "essential" genes and found that ~9% of these genetic perturbations could in fact be overcome by adaptive evolution. Our analyses uncovered a genome-wide gradient of gene essentiality, with certain essential cellular functions being more "evolvable" than others. Ploidy changes were prevalent among the evolved mutant strains, and aneuploidy of a specific chromosome was adaptive for a class of evolvable nucleoporin mutants. These data justify a quantitative redefinition of gene essentiality that incorporates both viability and evolvability of the corresponding mutant cells and will enable selection of therapeutic targets associated with lower risk of emergence of drug resistance.

In this work, we propose a novel Cycle In Cycle Generative Adversarial Network (C 2 GAN) for the task of keypoint-guided image generation. The proposed C 2 GAN is a cross-modal framework exploring a joint exploitation of the keypoint and the image data in an interactive manner. C 2 GAN contains two different types of generators, i.e., keypoint-oriented generator and image-oriented generator. Both of them are mutually connected in an end-to-end learnable fashion and explicitly form three cycled sub-networks, i.e., one image generation cycle and two keypoint generation cycles. Each cycle not only aims at reconstructing the input domain, and also produces useful output involving in the generation of another cycle. By so doing, the cycles constrain each other implicitly, which provides complementary information from the two different modalities and brings extra supervision across cycles, thus facilitating more robust optimization of the whole network. Extensive experimental results on two publicly available datasets, i.e., Radboud Faces [19] and Market-1501 [58], demonstrate that our approach is effective to generate more photo-realistic images compared with state-of-the-art models.

Schizosaccharomyces pombe is a widely used model organism to study many aspects of eukaryotic cell physiology. Its popularity as an experimental system partially stems from the ease of genetic manipulations, where the innate homology-targeted repair is exploited to precisely edit the genome. While vectors to incorporate exogenous sequences into the chromosomes are available, most are poorly characterized. Here, we show that commonly used fission yeast vectors, which upon integration produce repetitive genomic regions, give rise to unstable genomic loci. We overcome this problem by designing a new series of stable integration vectors (SIVs) that target four different prototrophy genes. SIVs produce non-repetitive, stable genomic loci and integrate predominantly as single copy. Additionally, we develop a set of complementary auxotrophic alleles that preclude false-positive integration events. We expand the vector series to include antibiotic resistance markers, promoters, fluorescent tags and terminators, and build a highly modular toolbox to introduce heterologous sequences. Finally, as proof of concept, we generate a large set of ready-to-use, fluorescent probes to mark organelles and cellular processes with a wide range of applications in fission yeast research.

Recently, head pose estimation (HPE) from low-resolution surveillance data has gained in importance. However, monocular and multi-view HPE approaches still work poorly under target motion, as facial appearance distorts owing to camera perspective and scale changes when a person moves around. To this end, we propose FEGA-MTL, a novel framework based on Multi-Task Learning (MTL) for classifying the head pose of a person who moves freely in an environment monitored by multiple, large field-of-view surveillance cameras. Upon partitioning the monitored scene into a dense uniform spatial grid, FEGA-MTL simultaneously clusters grid partitions into regions with similar facial appearance, while learning region-specific head pose classifiers. In the learning phase, guided by two graphs which a-priori model the similarity among (1) grid partitions based on camera geometry and (2) head pose classes, FEGA-MTL derives the optimal scene partitioning and associated pose classifiers. Upon determining the target's position using a person tracker at test time, the corresponding region-specific classifier is invoked for HPE. The FEGA-MTL framework naturally extends to a weakly supervised setting where the target's walking direction is employed as a proxy in lieu of head orientation. Experiments confirm that FEGA-MTL significantly outperforms competing single-task and multi-task learning methods in multi-view settings.

Abstract-Robust action recognition under viewpoint changes has received considerable attention recently. To this end, self-similarity matrices (SSMs) have been found to be effective view-invariant action descriptors. To enhance the performance of SSM-based methods, we propose multitask linear discriminant analysis (LDA), a novel multitask learning framework for multiview action recognition that allows for the sharing of discriminative SSM features among different views (i.e., tasks). Inspired by the mathematical connection between multivariate linear regression and LDA, we model multitask multiclass LDA as a single optimization problem by choosing an appropriate class indicator matrix. In particular, we propose two variants of graphguided multitask LDA: 1) where the graph weights specifying view dependencies are fixed a priori and 2) where graph weights are flexibly learnt from the training data. We evaluate the proposed methods extensively on multiview RGB and RGBD video data sets, and experimental results confirm that the proposed approaches compare favorably with the state-of-the-art.Index Terms-Multi-view action recognition, self-similarity matrix, multi-task learning, linear discriminant analysis.

BackgroundRecombinant protein production in the methylotrophic yeast Pichia pastoris largely relies on integrative vectors. Although the stability of integrated expression cassettes is well appreciated for most applications, the availability of reliable episomal vectors for this host would represent a useful tool to expedite cloning and high-throughput screening, ameliorating also the relatively high clonal variability reported in transformants from integrative vectors caused by off-target integration in the P. pastoris genome. Recently, heterologous and endogenous autonomously replicating sequences (ARS) were identified in P. pastoris by genome mining, opening the possibility of expanding the available toolbox to include efficient episomal plasmids. The aim of this technical report is to validate a 452-bp sequence (“panARS”) in context of P. pastoris expression vectors, and to compare their performance to classical integrative plasmids. Moreover, we aimed to test if such episomal vectors would be suitable to sustain in vivo recombination, using fragments for transformation, directly in P. pastoris cells.ResultsA panARS-based episomal vector was evaluated using blue fluorescent protein (BFP) as a reporter gene. Normalized fluorescence from colonies carrying panARS-BFP outperformed the level of signal obtained from integrative controls by several-fold, whereas endogenous sequences, identified from the P. pastoris genome, were not as efficient in terms of protein production. At the single cell level, panARS-BFP clones showed lower interclonal variability but higher intraclonal variation compared to their integrative counterparts, supporting the idea that heterologous protein production could benefit from episomal plasmids. Finally, efficiency of 2-fragment and 3-fragment in vivo recombination was tested using varying lengths of overlapping regions and molar ratios between fragments. Upon optimization, minimal background was obtained for in vivo assembled vectors, suggesting this could be a quick and efficient method to generate of episomal plasmids of interest.ConclusionsAn expression vector based on the panARS sequence was shown to outperform its integrative counterparts in terms of protein productivity and interclonal variability, facilitating recombinant protein expression and screening. Using optimized fragment lengths and ratios, it was possible to perform reliable in vivo recombination of fragments in P. pastoris. Taken together, these results support the applicability of panARS episomal vectors for synthetic biology approaches.

The cohesin ring, which is composed of the Smc1, Smc3, and Scc1 subunits, topologically embraces two sister chromatids from S phase until anaphase to ensure their precise segregation to the daughter cells. The opening of the ring is required for its loading on the chromosomes and unloading by the action of Wpl1 protein. Both loading and unloading are dependent on ATP hydrolysis by the Smc1 and Smc3 "head" domains, which engage to form two composite ATPase sites. Based on the available structures, we modeled the Saccharomyces cerevisiae Smc1/Smc3 head heterodimer and discovered that the Smc1/Smc3 interfaces at the two ATPase sites differ in the extent of protein contacts and stability after ATP hydrolysis. We identified smc1 and smc3 mutations that disrupt one of the interfaces and block the Wpl1-mediated release of cohesin from DNA. Thus, we provide structural insights into how the cohesin heads engage with each other.

Abstract-