Large Language Models (LLMs) are widely used as conversational agents, exploiting their capabilities in various sectors such as education, law, medicine, and more.However, LLMs are often subjected to context-shifting behaviour, resulting in a lack of consistent and interpretable personality-aligned interactions.Adherence to psychological traits lacks comprehensive analysis, especially in the case of dyadic (pairwise) conversations.We examine this challenge from two viewpoints, initially using two conversation agents to generate a discourse on a certain topic with an assigned personality from the OCEAN framework (Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism) as High/Low for each trait.This is followed by using multiple judge agents to infer the original traits assigned to explore prediction consistency, intermodel agreement, and alignment with the assigned personality.Our findings indicate that while LLMs can be guided toward personalitydriven dialogue, their ability to maintain personality traits varies significantly depending on the combination of models and discourse settings.These inconsistencies emphasise the challenges in achieving stable and interpretable personality-aligned interactions in LLMs.

Precise localization is critical for autonomous vehicles. We present a self-supervised learning method that employs transformers for the first time for the task of outdoor localization using LiDAR data. We propose a pre-text task that reorganizes the slices of a 360° LiDAR scan to leverage its axial properties. Our model, called Slice Transformer, employs multi-head attention while systematically processing the slices. To the best of our knowledge, this is the first instance of leveraging multi-head attention for outdoor point clouds. We additionally introduce the Perth-Wadataset, which provides a large-scale LiDAR map of Perth city in Western Australia, covering ~4km2area. Localization annotations are provided for Perth - Wa.The proposed localization method is thoroughly evaluated on Perth-WA and Appollo-SouthBay datasets. We also establish the efficacy of our self-supervised learning approach for the common downstream task of object classification using ModelNet40 and ScanNN datasets. The code and Perth-WA data will be publicly released.

MOTIVATION: The secondary structure of RNA is of importance to its function. Over the last few years, several papers attempted to use machine learning to improve de novo RNA secondary structure prediction. Many of these papers report impressive results for intra-family predictions but seldom address the much more difficult (and practical) inter-family problem. RESULTS: We demonstrate that it is nearly trivial with convolutional neural networks to generate pseudo-free energy changes, modelled after structure mapping data that improve the accuracy of structure prediction for intra-family cases. We propose a more rigorous method for inter-family cross-validation that can be used to assess the performance of learning-based models. Using this method, we further demonstrate that intra-family performance is insufficient proof of generalization despite the widespread assumption in the literature and provide strong evidence that many existing learning-based models have not generalized inter-family. AVAILABILITY AND IMPLEMENTATION: Source code and data are available at https://github.com/marcellszi/dl-rna. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Metagenomics is the sequencing and study of DNA directly from environmental samples and is an approach used in microbial ecology to explore the diversity and functions of microbial communities [1].The study of the metagenome presents an avenue to investigate environmental microorganisms in-situ through culture-independent techniques, which facilitates the discovery and analysis of the metabolic potential of hitherto uncultured taxa [2].Expanding our reach into the once obscure facets of microbial life has yielded further insights into the ecological activity of microorganisms, uncovering a trove of novel metabolic products and pathways with a wide range of biotechnical applications in many industries [3,4].Data obtained from metagenomes have also been pivotal in directing strategies for the cultivation of novel microorganisms [5].The benefit of metagenomics is that it forms a wide-reaching preliminary approach that can provide contextual information to direct further downstream research for microorganisms in the wet lab.Microbial ecology has developed to a stage where "big data" is often the norm.Through advances in sequencing and computational technologies, we have experienced an unprecedented level of progress in the field of microbial ecology, improving our understanding of the many global ecosystems [6].Metagenomics draws many parallels when compared to data science-environmental sequence data presents a large volume of information that must be sifted through and sorted to make sense of underlying ecological and taxonomic patterns [7].The amount of information that needs to be processed, however, can be daunting for those starting out and often still pose challenges even for those familiar to the field.The inspiration behind this writing is from a recommendation to a PLOS Ten Simple Rules article on command-line bioinformatics [8].The article provided a lot of information on tools and concepts used by bioinformaticians and it would have been a valuable source of guidance for many.This paper styled in the typical format of a Ten Simple Rules article, presents a guide on the basics of metagenomic-based research aimed at environmental settings.We will explore the concepts, tools, and resources utilized in metagenomics and how these can potentially help the progress of a researcher beginning their studies (Fig 1).The 10 simple rules format is a casual one and people are often busy, so at the end of each rule is a TL; DR (too long; didn't read) that summarizes the point we are trying to make if you want to skim through (although we do hope you'll stick around and read the paper in depth!).

MOTIVATION: Predicting the secondary structure of RNA is a fundamental task in bioinformatics. Algorithms that predict secondary structure given only the primary sequence, and a model to evaluate the quality of a structure, are an integral part of this. These algorithms have been updated as our model of RNA thermodynamics changed and expanded. An exception to this has been the treatment of multi-loops. Although more advanced models of multi-loop free energy change have been suggested, a simple, linear model has been used since the 1980s. However, recently, new dynamic programing algorithms for secondary structure prediction that could incorporate these models were presented. Unfortunately, these models appear to have lower accuracy for secondary structure prediction. RESULTS: We apply linear regression and a new parameter optimization algorithm to find better parameters for the existing linear model and advanced non-linear multi-loop models. These include the Jacobson-Stockmayer and Aalberts & Nandagopal models. We find that the current linear model parameters may be near optimal for the linear model, and that no advanced model performs better than the existing linear model parameters even after parameter optimization. AVAILABILITY AND IMPLEMENTATION: Source code and data is available at https://github.com/maxhwardg/advanced_multiloops. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

This paper presents a preliminary study into the awarding of barnstars among Wikipedia editors to better understand their motivations in contributing to Wikipedia articles. We crawled the talk pages of all active Wikipedia editors and retrieved 21,299 barnstars that were awarded among 14,074 editors. In particular, we found that editors do not award and receive barnstars in equal (or similar) quantities. Also, editors were more active in editing articles before awarding or receiving barnstars.

MOTIVATION: Laboratory RNA structure determination is demanding and costly and thus, computational structure prediction is an important task. Single sequence methods for RNA secondary structure prediction are limited by the accuracy of the underlying folding model, if a structure is supported by a family of evolutionarily related sequences, one can be more confident that the prediction is accurate. RNA pseudoknots are functional elements, which have highly conserved structures. However, few comparative structure prediction methods can handle pseudoknots due to the computational complexity. RESULTS: A comparative pseudoknot prediction method called DotKnot-PW is introduced based on structural comparison of secondary structure elements and H-type pseudoknot candidates. DotKnot-PW outperforms other methods from the literature on a hand-curated test set of RNA structures with experimental support. AVAILABILITY: DotKnot-PW and the RNA structure test set are available at the web site http://dotknot.csse.uwa.edu.au/pw. CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

BACKGROUND: The late embryogenesis abundant (LEA) proteins cover a number of loosely related groups of proteins, originally found in plants but now being found in non-plant species. Their precise function is unknown, though considerable evidence suggests that LEA proteins are involved in desiccation resistance. Using a number of statistically-based bioinformatics tools the classification of a large set of LEA proteins, covering all Groups, is reexamined together with some previous findings. Searches based on peptide composition return proteins with similar composition to different LEA Groups; keyword clustering is then applied to reveal keywords and phrases suggestive of the Groups' properties. RESULTS: Previous research has suggested that glycine is characteristic of LEA proteins, but it is only highly over-represented in Groups 1 and 2, while alanine, thought characteristic of Group 2, is over-represented in Group 3, 4 and 6 but under-represented in Groups 1 and 2. However, for LEA Groups 1 2 and 3 it is shown that glutamine is very significantly over-represented, while cysteine, phenylalanine, isoleucine, leucine and tryptophan are significantly under-represented. There is also evidence that the Group 4 LEA proteins are more appropriately redistributed to Group 2 and Group 3. Similarly, Group 5 is better found among the Group 3 LEA proteins. CONCLUSIONS: There is evidence that Group 2 and Group 3 LEA proteins, though distinct, might be related. This relationship is also evident in the overlapping sets of keywords for the two Groups, emphasising alpha-helical structure and, at a larger scale, filaments, all of which fits well with experimental evidence that proteins from both Groups are natively unstructured, but become structured under stress conditions. The keywords support localisation of LEA proteins both in the nucleus and associated with the cytoskeleton, and a mode of action similar to chaperones, perhaps the cold shock chaperones, via a role in DNA-binding. In general, non-globular and low-complexity proteins, such as the LEA proteins, pose particular challenges in determining their functions and modes of action. Rather than masking off and ignoring low-complexity domains, novel tools and tool combinations are needed which are capable of analysing such proteins in their entirety.

Abstract The POPPs is a suite of inter-related software tools which allow the user to discover what is statistically ‘unusual’ in the composition of an unknown protein, or to automatically cluster proteins into families based on peptide composition. Finally, the user can search for related proteins based on peptide composition. Statistically based peptide composition provides a view of proteins that is, to some extent, orthogonal to that provided by sequence. In a test study, the POPP suite is able to regroup into their families sets of approximately 100 randomised Pfam protein domains. The POPPs suite is used to explore the diverse set of late embryogenesis abundant (LEA) proteins. Availability: Contact the author. Contact: [email protected] Keywords: protein clustering; clustering algorithms; protein search; peptide composition; Pfam; late embryogenesis abundant proteins.

Terminal-restriction fragment length polymorphism (T-RFLP) and length heterogeneity-polymerase chain reaction (LH-PCR) are DNA fingerprinting technologies which use PCR amplification of a gene of interest e.g. small subunit rRNA gene, to study microbial community structure and dynamics. Either one or both of the forward and reverse strand primers used to amplify the gene are fluorescently labelled. The products of restriction endonuclease digestion are electrophoresed with an automated sequencer that detects only the terminal (labelled) restriction fragments (T-RFs). In LH-PCR, products are electrophoresed without digestion, with different fragment lengths being due to inherent variation in the amplified sequence. A novel software system, TRUFFLER, has been developed to mimic this process in silico allowing comparison of experimental data against databases of theoretically determined T-RFs. As a given combination of forward and reverse primers and restriction endonuclease can yield identical T-RFs across a number of species, combinations of different endonucleases (and/or primers) are typically used In addition to fragment length data, data on fluorescence levels is also available. Computationally, this can be viewed as a constraint satisfaction problem which can be solved to allow identification of the dominant members of the microbial community, often down to individual species or at least genus level, and their relative proportions.

The Protein Annotators' Assistant (or PAA) (http://www.ebi.ac.uk/paa/) is a software system which assists protein annotators in the task of assigning functions to newly sequenced proteins. Working backward from SwissProt, a database which describes known proteins, and a prior sequence similarity search that returns a list of known proteins similar to a query, PAA suggests keywords and phrases which may describe functions performed by the query. In a preprocessing step, a database is built from the protein names that appear in the SwissProt database, and against each protein are listed key words and phrases that are extracted from the corresponding text records. Common words either in general English usage or from the biological domain are removed as the phrases are assembled. This process is assisted by the use of a simple stemming algorithm, which extends the list of stop-words (i.e., reject words), together with a list of accept-words. At runtime, the search algorithm, invoked by a user via a Web interface, takes a list of protein names and clusters the named proteins around keywords/phrases shared by members of the list. The assumption is that if these proteins have a particular keyword/phrase in common, and they are related to a query protein, then the keyword/phrase may also describe the query. Overall, PAA employs a number of IR techniques in a novel setting and is thus related to text categorization, where multiple categories may be suggested, except that in this case none of the categories are specified in advance.

Traditional programming environments represent program source code as a set of source files. These files have various ‘dependencies’ on each other, such that a file needs recompilation if it depends on a file which has changed. A ‘build tool’ is used to process these dependencies and bring the application ‘up-to-date’. An example of a build tool is the UNIX ‘make’. This paper examines what happens when the dependencies used by the build tool are expressed between functions (or objects) rather than between files. Qualitative differences arise from the difference in granularity. The result is an effective incremental compilation programming environment, based on the C++ language. It is called ‘Barbados’, and is fully implemented. The environment resembles an interpreter in that changes to source code appear to be immediately reflected in all object code, except that errors are reported early as in compiled systems. Incremental compilation is not a well-used technology, possibly because the ‘fine-grain build’ problem is not well understood. Nevertheless, incremental compilation systems do exist. The advantages of the system described here are that it works with relatively standard compilation technology, it works for the C++ language including the preprocessor, it is an elegant solution and it is more efficient than competing algorithms. © 1997 by John Wiley & Sons, Ltd.

In spite of years of effort, plagiarism in student assignment submissions still causes considerable difficulties for course designers; if students' work is not their own, how can anyone be certain they have learnt anything? YAP is a system for detecting suspected plagiarism in computer programs and other texts submitted by students. The paper reviews YAP3, the third version of YAP, focusing on its novel underlying algorithm - Running-Karp-Rabin Greedy-String-Tiling (or RKS-GST), whose development arose from the observation with YAP and other systems that students shuffle independent code segments. YAP3 is able to detect transposed subsequences, and is less perturbed by spurious additional statements. The paper concludes with a discussion of recent extension of YAP to English texts, further illustrating the flexibility of the YAP approach.

Early automated systems for detecting plagiarism in student programs employed attribute counting techniques in their comparisons of program texts, while more recent systems use encoded structural information. Whales claims that the latter are more effective in their detection of plagiarisms than systems based on attribute counting. To explore the validity of these claims, a comparison is presented of five systems, two based on attribute counting and three using metrics based on structure. The major result of this study is that the systems based on structural information consistently equal or better the performance of systems based on attribute counting metrics. A second conclusion is that of the structure metric systems, one using approximate tokenization of input texts (YAP) is as effective as a system that undertakes a complete parse (Plague). Approximate tokenization offers a considerable reduction in the costs of porting to new languages. A distinction is also made between forms of plagiarism common among novice programmers and those employed by more experienced programmers.

Abstract PMS‐Prolog is an implementation of Prolog designed for distributed systems, and therefore uses coarsegrain parallelism in preference to the fine‐grain parallelism of other implementations. Prolog processes are formally defined and interprocess communication is explicit. An overview of PMS‐Prolog is provided, followed by an examination of the PMS‐Prolog solutions to four problems. These solutions are also compared to other solutions from the literature. From this, the strengths and weaknesses of the PMS‐Prolog model can be seen. Avenues for improving the model are then suggested.

It has argued that systems for detecting similarities in submitted programs must be based on the analysis of program structure. One such system is “Plague”. In this paper, some difficulties with the Plague system are outlined and a new system is described that achieves the same goals as Plague, but in a way that is both more portable and more easily customized to new target languages.

The EPILOG language and model of computation are briefly described, together with four multiprocessor architectures on which it is proposed to run the model: a form of chordal network and a slight variant of Wu and Feng's Baseline and Reverse Baseline networks. Results are reported for the simulated execution of some EPILOG programs on these architectures. Finally, some tentative conclusions are drawn and some possible avenues for future research are indicated.