Medical domain knowledge in domain-agnostic generative AI

The text-guided diffusion model GLIDE (Guided Language to Image Diffusion for Generation and Editing) is the state of the art in text-to-image generative artificial intelligence (AI). GLIDE has rich representations, but medical applications of this model have not been systematically explored. If GLIDE had useful medical knowledge, it could be used for medical image analysis tasks, a domain in which AI systems are still highly engineered towards a single use-case. Here we show that the publicly available GLIDE model has reasonably strong representations of key topics in cancer research and oncology, in particular the general style of histopathology images and multiple facets of diseases, pathological processes and laboratory assays. However, GLIDE seems to lack useful representations of the style and content of radiology data. Our findings demonstrate that domain-agnostic generative AI models can learn relevant medical concepts without explicit training. Thus, GLIDE and similar models might be useful for medical image processing tasks in the future - particularly with additional domain-specific fine-tuning.

The human RNA polymerase I structure reveals an HMG-like transcription factor docking domain specific to metazoans

Transcription of the ribosomal RNA precursor by RNA polymerase (Pol) I is a major determinant of cellular growth and dysregulation is observed in many cancer types. Here, we present the purification of human Pol I from cells carrying a genomic GFP-fusion on the largest subunit allowing the structural and functional analysis of the enzyme across species. In contrast to yeast, human Pol I carries a single-subunit stalk and in vitro transcription indicates a reduced proofreading activity. Determination of the human Pol I cryo-EM reconstruction in a close-to-native state rationalizes the effects of disease-associated mutations and uncovers an additional domain that is built into the sequence of Pol I subunit RPA1. This "dock II" domain resembles a truncated HMG-box incapable of DNA-binding which may serve as a downstream-transcription factor binding platform in metazoans. Biochemical analysis and ChIP data indicate that Topoisomerase 2a can be recruited to Pol I via the domain and cooperates with the HMG-box domain containing factor UBF. These adaptations of the metazoan Pol I transcription system may allow efficient release of positive DNA supercoils accumulating downstream of the transcription bubble.

Collision Avoidance Algorithm for USV Based on Rolling Obstacle Classification and Fuzzy Rules

Dynamic collision avoidance between multiple vessels is a task full of challenges for unmanned surface vehicle (USV) movement, which has high requirements on real-time performance and safety. The difficulty of multi-obstacle collision avoidance is that it is hard to formulate the optimal obstacle avoidance strategy when encountering more than one obstacle threat at the same time; a good strategy to avoid one obstacle sometimes leads to threats from other obstacles. This paper presents a dynamic collision avoidance algorithm for USVs based on rolling obstacle classification and fuzzy rules. Firstly, potential collision probabilities between a USV and obstacles are calculated based on the time to the closest point of approach (TCPA). All obstacles are given different priorities based on potential collision probability, and the most urgent and secondary urgent ones will then be dynamically determined. Based on the velocity obstacle algorithm, four possible actions are defined to determine the basic domain in the collision avoidance strategy. After that, the Safety of Avoidance Strategy and Feasibility of Strategy Adjustment are calculated to determine the additional domain based on fuzzy rules. Fuzzy rules are used here to comprehensively consider the situation composed of multiple motion obstacles and the USV. Within the limited range of the basic domain and the additional domain, the optimal collision avoidance parameters of the USV can be calculated by the particle swarm optimization (PSO) algorithm. The PSO algorithm utilizes both the characteristic of pursuance for the population optimal and the characteristic of exploration for the individual optimal to avoid falling into the local optimal solution. Finally, numerical simulations are performed to certify the validity of the proposed method in complex traffic scenarios. The results illustrated that the proposed method could provide efficient collision avoidance actions.

Free Space Detection Using Camera-LiDAR Fusion in a Bird’s Eye View Plane

Although numerous road segmentation studies have utilized vision data, obtaining robust classification is still challenging due to vision sensor noise and target object deformation. Long-distance images are still problematic because of blur and low resolution, and these features make distinguishing roads from objects difficult. This study utilizes light detection and ranging (LiDAR), which generates information that camera images lack, such as distance, height, and intensity, as a reliable supplement to address this problem. In contrast to conventional approaches, additional domain transformation to a bird’s eye view space is executed to obtain long-range data with resolutions comparable to those of short-range data. This study proposes a convolutional neural network architecture that processes data transformed to a bird’s eye view plane. The network’s pathways are split into two parts to resolve calibration errors in the transformed image and point cloud. The network, which has modules that operate sequentially at various scaled dilated convolution rates, is designed to quickly and accurately handle a wide range of data. Comprehensive empirical studies using the Karlsruhe Institute of Technology and Toyota Technological Institute’s (KITTI’s) road detection benchmarks demonstrate that this study’s approach takes advantage of camera and LiDAR information, achieving robust road detection with short runtimes. Our result ranks 22nd in the KITTI’s leaderboard and shows real-time performance.

Archaeal origins of gamete fusion

Sexual reproduction in Eukarya consists of genome reduction by meiosis and subsequent gamete fusion. The presence of meiotic genes in Archaea and Bacteria suggests that prokaryotic DNA repair mechanisms evolved towards meiotic recombination. However, the evolutionary origin of gamete fusion is less clear because fusogenic proteins resembling those found in Eukarya have so far not been identified in prokaryotes. Here, using bioinformatics, we identified archaeal genes encoding candidates of fusexins, a superfamily of fusogens mediating somatic and gamete fusion in multiple eukaryotic lineages. Crystallographic structure determination of a candidate archaeal FusexinA reveals an archetypical trimeric fusexin architecture with novel features such as a six-helix bundle and an additional globular domain. We demonstrate that ectopically expressed FusexinA can fuse mammalian cells, and that this process involves the additional domain and a more broadly conserved fusion loop. Genome content analyses reveal that archaeal fusexins genes are within integrated mobile elements. Finally, evolutionary analyses place these archaeal fusogens as the founders of the fusexin superfamily. Based on these findings, we propose a new hypothesis on the origins of eukaryotic sex where an archaeal fusexin, originally used by selfish elements for horizontal transmission, was repurposed to enable gamete fusion.

Cybersecurity Verification and Validation Testing in Automotive

The new generations of cars have a number of ECUs (Electronic Control Units) which are connected to a central gateway and need to pass cybersecurity integration tests to fulfil the homologation requirements of cars. Cars usually have a gateway server (few have additional domain servers) with Linux and a large number of ECUs which are real time control of actuators (ESP, EPS, ABS, etc. – usually they are multicore embedded controllers) connected by a real time automotive specific bus (CAN-FD) to the domain controller or gateway server. The norms (SAE J3061, ISO 21434) require cybersecurity related verification and validation. Fir the verification car manufacturers use a network test suite which runs > 2000 test cases and which have to be passed for homologation. These norms have impact on the way how car communication infrastructure is tested, and which cybersecurity attack patterns are checked before a road release of an ECU/car. This paper describes typical verification and validation approaches in modern vehicles and how such test cases are derived and developed.

Domains of Female Choice in Human Evolution

The present article advances the view that women’s mate preferences can be grouped into at least two overarching domains: competitiveness and fatherhood. Theoretical and empirical considerations suggest that female mate preferences evolve in contexts of male competitiveness and often amplify the effects of male-male competition. Evidence for the importance of male-male competition and female choice for competitiveness in humans is reviewed. Evidence is likewise offered for the importance of human fatherhood as an additional domain of female choice outside of male competitiveness. Implications of more inclusive mate preferences for the evolution of cognitive architecture are discussed.

Lower Kolyma multilingualism: Historical setting and sociolinguistic trends

Aims and objectives/purpose/research questions: The region of Lower Kolyma (LK), Russia, exhibits a unique multilingual situation which involves five languages from five different families: Even (Tungusic), Yakut (Turkic), Chukchi (Chukotko-Kamchatkan), Tundra Yukaghir (Yukaghir), and Russian (Slavic, Indo-European). Our research aims to show the ongoing changes in this setting and to explain the historical background of these trends. Design/methodology/approach: Our approach combines the results of a sociolinguistic survey, published data about the history and ethnography of the region, and observed linguistic effects of language contact. Data and analysis: The sample for our sociolinguistic study includes 196 individuals, born between 1878 and 1996. To assess the patterns of multilingualism we use two parameters: the number of acquired languages and the composition of the linguistic repertoires. To model the linguistic situation in the 20th century, an average life span parameter is used. Findings/conclusions: We demonstrate that in 1940 to 1980 it was most common for LK residents to use all languages of the region. Partly, at least, this was a consequence of the introduction of boarding schools, which became an additional domain for interethnic communication. In recent decades we observe a tendency for a shift to Russian with a concomitant decrease in speakers of the minority languages. Originality: We use an approach of retrospective interviews which allows us to cover a large time period and to provide an overview of changes in the multilingual setting during the 20th century. Thus, even though multilingualism in this area has already been discussed by Vakhtin, the methodological approach is different. Significance/implications: The paper contributes to the field of sociolinguistics, presenting a new case study of multilingualism from an area of Siberia which is understudied in this regard.

A comparative analysis of information gathering meetings conducted by novice design teams across multiple design project stages

Designers often gather information, for instance through stakeholder or domain expert meetings, to understand their design problems and develop effective solutions. However, few previous studies have provided in-depth descriptions of novice engineering designers' approaches to conducting information gathering meetings. In this preliminary study, we analyzed data from six capstone mechanical engineering design teams to identify the types of individuals from whom teams gathered information, when these meetings occurred, and how teams solicited information during meetings. Teams in our study demonstrated a range of information gathering behaviors that aligned with recommended practices, particularly in their early meetings. Furthermore, we observed relatively few instances of teams exhibiting behaviors that were less similar to recommended practices during meetings. However, our findings also revealed two key trends across teams that represented specific opportunities for improvement and that may reflect novice approaches to conducting information gathering meetings. First, teams consistently explored domain experts' perspectives in depth during meetings but did not typically explore the perspectives of their project partners in similar depth. Teams also met with additional domain experts to inform their projects. In addition, teams seemed to finalize their design decisions in their early design meetings and were less likely to conduct information gathering meetings in later design phases. The comprehensive descriptions of novice mechanical engineering designers' approaches provided in our preliminary study provide an entry point for further investigations that can inform engineering training, tools, and pedagogy for conducting effective information gathering meetings.