Structure and genetic variability of the Mexican Sardo Negro breed

ABSTRACT: This study analyzed the Sardo Negro breed pedigree (41,521 animals registered from 1958 to 2019) to determine its structure, evolution, and genetic variability (GV). The population genetic parameters evaluated were effective number of founders (fe) and ancestors (fa), pedigree integrity, additive genetic relationship (AGR); number of complete generations (NCG), maximum generations traced (NMGT), and equivalent complete generations (NECG); effective population size (Ne), inbreeding coefficient (F), and generation interval (GI). The average GI was 7.45 years. A total of 7,804 founders and 4,856 ancestors were identified for a fe of 185 and a fa of 97. The average and maximum values of NCG, NECG, and NMGT were 1.6 and 5.0, 2.5 and 6.5, 4.3 and 12, with Ne estimates of 15.9, 25.9, and 69.0, respectively. The increase in F, linked to Ne, ranged from 0.72% to 3.1% per generation. The average values for F and AGR were 3.6% and 1.0%, respectively. The proportion of inbred individuals was 32.0%, with F values ranging from 0.01 to 62.2% and an average of 11.3%. The rate of inbred population was 1.3% per year. The annual rate of AGR was 0.04%. For the continuity and projection of the breed, the evolution of F as a function of Ne and the possible implications of the selection schemes must be considered. The genetic variability sustained over time results from the Ne.

Design of a 9-bit column-parallel ADC in the MAPS for real-time beam monitoring

This paper describes a 2 Msps 9-bit column-parallel ADC for monolithic active pixel sensor. It is designed in fully differential cyclic architecture and takes eight clock cycles to perform a 9-bit conversion. This ADC is fabricated in a 130 nm CMOS process. Each ADC covers a small area of 100 µm × 300 µm and consumes ∼5 mW. The measurement results show that this ADC has a signal-to-noise and distortion ratio (SNDR) of 46.8 dB. The DNL (Differential Nonlinearity) and (Integral Nonlinearity) INL are 0.168 LSB and 0.112 LSB, respectively. The effective number of bits (ENOB) is 7.48 bits.

Genealogical analyses in open population of Silla Argentino horses belonging to the national police of Colombia

This study aimed to evaluate population parameters, genetic diversity and identify the ancestors with the highest genetic contribution of an open population of Silla Argentino horses in Colombia using pedigree data. 508 horses born between 1998 and 2018 were evaluated, called reference population (RP). The ancestors of the RP were added, defining the total population (TP) whit 1,861 horses born since 1905. Population parameters were estimated, including inbreeding coefficient (F), effective population size (Ne), effective number of founders (fe), ancestors (fa), and founder genomes (fg) and generation interval (GIN). The GIN was 10.1 (TP) and 7.7 (RP) years. The estimate of F for the RP was 0.3%, indicating control in mattings. The Ne was 317 (RP) and 179 (TP). The probability of genes origin indicated values for fe, fa and fg of 386, 187 and 351 (TP), and 161, 64 and 56 (RP), respectively, evidencing a reduction of genetic variability in the RP. Half of the genetic variation in the equine population studied was explained by 40 ancestors. The results show loss of genetic diversity in the RP and the genetic contribution of the ancestors suggests the need to increase the number of stallions for breeding and guide mating to increase genetic diversity in the new generations of Silla Argentino horses of the Colombian National Police

Genetic diversity, viability and conservation value of the global captive population of the Moroccan Royal lions

This study evaluates the diversity of the so-called ‘Moroccan Royal lions’ using genealogical information. Lions are no longer extant in North Africa, but the previous wild population was an important element of the now-recognised northern subspecies (Panthera leo leo) that ranged across West Africa, North Africa and the Middle East into India. The remaining captive population of ‘Moroccan Royal lions’ seems to be significantly endangered by the loss of diversity due to the effective population size decrease. The pedigree file of this captive lion population consisted of 454 individuals, while the reference population included 98 animals (47 males and 51 females). The completeness of the pedigree data significantly decreased with an increasing number of generations. The highest percentage of pedigree completeness (over 70%) was achieved in the first generation of the reference population. Pedigree-based parameters derived from the common ancestor and gene origin were used to estimate the state of diversity. In the reference population, the average inbreeding coefficient was 2.14%, while the individual increase in inbreeding over generations was 2.31%. Overall, the reference population showed lower average inbreeding and average relatedness compared with the pedigree file. The number of founders (47), the effective number of founders (24) and the effective number of ancestors (22) were estimated in the reference population. The effective population size of 14.02 individuals confirms the critically endangered status of the population and rapid loss of diversity in the future. Thus, continuous monitoring of the genetic diversity of the ‘Moroccan Royal lion’ group is required, especially for long-term conservation management purposes, as it would be an important captive group should further DNA studies establish an affinity to P. leo leo.

A Novel Method for Goal Recognition from 10 m Distance Using Deep Learning in CanSat

In this study, we propose a method for CanSat to recognize and guide a goal using deep learning image classification even 10 m away from the goal, and describe the results of demonstrative evaluation to confirm the effectiveness of the method. We applied deep learning image classification to goal recognition in CanSat for the first time at ARLISS 2019, and succeeded in guiding it almost all the way to the goal in all three races, winning the first place as overall winner. However, the conventional method has a drawback in that the goal recognition rate drops significantly when the CanSat is more than 6–7 m away from the goal, making it difficult to guide the CanSat to the goal when it moves away from the goal because of various factors. To enable goal recognition from a distance of 10 m from the goal, we investigated the number of horizontal regions of interest divisions and the method of vertical shifts during image recognition, and clarified the effective number of divisions and recognition rate using experiments. Although object detection is commonly used to detect the position of an object from an image by deep learning, we confirmed that the proposed method has a higher recognition rate at long distances and a shorter computation time than SSD MobileNet V1. In addition, we participated in the CanSat contest ACTS 2020 to evaluate the effectiveness of the proposed method and achieved the zero-distance goal in all three competitions, demonstrating its effectiveness by winning first place in the comeback category.

Ewolucja systemu partyjnego i sceny politycznej w Bułgarii w latach 1990-2017 – od dwublokowości do wielopartyjności

This article concerns the evolution of the party system and the political scene of Bulgaria after the system transformations. The foundations of this analysis are the results of the elections to the National Assembly in the years 1990-2017. In order to identify the type of the party system, official election results and the effective number of parties index (ENP) were used. This index was used to characterise the party system after 2001. Some conditions from communism time had an influence on the shape and kind of how the Bulgarian party system and political scene evolved after system transformation. Between 1990-2001 in Bulgaria, a two-block or two and a half-block party system existed. From 2001 it evolves in the direction of a multiparty system.

Long-tailed Visual Recognition via Gaussian Clouded Logit Adjustment

<p>Long-tailed data is still a big challenge for deep neural networks, even though they have achieved great success on balanced data. We observe that vanilla training on long-tailed data with cross-entropy loss makes the instance-rich head classes severely squeeze the spatial distribution of the tail classes, which leads to difficulty in classifying tail class samples. Furthermore, the original cross-entropy loss can only propagate gradient short-lively because the gradient in softmax form rapidly approaches zero as the logit difference increases. This phenomenon is called softmax saturation. It is unfavorable for training on balanced data, but can be utilized to adjust the validity of the samples in long-tailed data, thereby solving the distorted embedding space of long-tailed problems. To this end, this paper therefore proposes the Gaussian clouded logit adjustment by Gaussian perturbing different class logits with varied amplitude. We define the amplitude of perturbation as cloud size and set relatively large cloud sizes to tail classes. The large cloud size can reduce the softmax saturation and thereby making tail class samples more active as well as enlarging the embedding space. To alleviate the bias in the classifier, we accordingly propose the class-based effective number sampling strategy with classifier re-training. Extensive experiments on benchmark datasets validate the superior performance of the proposed method.</p><br>

Long-tailed Visual Recognition via Gaussian Clouded Logit Adjustment

<p>Long-tailed data is still a big challenge for deep neural networks, even though they have achieved great success on balanced data. We observe that vanilla training on long-tailed data with cross-entropy loss makes the instance-rich head classes severely squeeze the spatial distribution of the tail classes, which leads to difficulty in classifying tail class samples. Furthermore, the original cross-entropy loss can only propagate gradient short-lively because the gradient in softmax form rapidly approaches zero as the logit difference increases. This phenomenon is called softmax saturation. It is unfavorable for training on balanced data, but can be utilized to adjust the validity of the samples in long-tailed data, thereby solving the distorted embedding space of long-tailed problems. To this end, this paper therefore proposes the Gaussian clouded logit adjustment by Gaussian perturbing different class logits with varied amplitude. We define the amplitude of perturbation as cloud size and set relatively large cloud sizes to tail classes. The large cloud size can reduce the softmax saturation and thereby making tail class samples more active as well as enlarging the embedding space. To alleviate the bias in the classifier, we accordingly propose the class-based effective number sampling strategy with classifier re-training. Extensive experiments on benchmark datasets validate the superior performance of the proposed method.</p><br>

The dimensionality of plant–plant competition

To avoid extinction, every species must be able to exploit available resources at least as well as the other species in its community. All else being equal, theory predicts that the more distinct the niches of such co-occurring and competing species, the more species that can persist in the long run. However, both theoretical and experimental studies define a priori the nature and number of resources over which species compete. It therefore remains unclear whether or not species in empirically realistic contexts are actually exploiting all or some of the niches available to them. Here we provide a mathematical solution to this long-standing problem. Specifically, we show how to use the interactions between sets of co-occurring plant species to quantify their implied "niche dimensionality": the effective number of resources over which those species appear to be competing. We then apply this approach to quantify the niche dimensionality of 12 plant assemblages distributed across the globe. Contrary to conventional wisdom, we found that the niche dimensionality in these systems was much lower than the number of competing species. However, two high-resolution experiments also show that changes in the local environment induce a reshuffling of plant's competitive roles and hence act to increase the assemblages' effective niche dimensionality. Our results therefore indicate that homogeneous environments are unlikely to be able to maintain high diversity and also shows how environmental variation impacts species' niches and hence their opportunities for long-term survival.