Expert systems in histopathology. II. Knowledge representation and rule-based systems

Two aspects of expert systems for use in diagnostic histopathology and cytopathology are examined: knowledge representation and the structure and operation of rule-based systems. Knowledge may be represented, e.g., in semantic networks, frames, multiple contexts and model-based structures; the choice of structure should be matched to the type of information to create an efficient and logically adequate expert system. In a rule-based system, knowledge is represented as "rules," often in the form of "IF (condition)-THEN (conclusion)" rules. The anatomy of such rules and their operation is explored via the use of examples. Uncertainty in rules is briefly addressed, and their processing by the symbolic reasoning of the "inference engine" of the expert system is described, including both "forward-chaining" ("data-driven") operations and "backward-chaining" ("goal-driven") operations.     

The ring between ring fingers (RBR) protein family

Proteins of the ring between ring fingers (RBR)-domain family are characterized by three groups of specifically clustered (typically eight) cysteine and histidine residues. Whereas the amino-terminal ring domain (N-RING) binds two zinc ions and folds into a classical cross-brace ring finger, the carboxy-terminal ring domain (C-RING) involves only one zinc ion. The three-dimensional structure of the central ring domain, the IBR domain, is still unsolved. About 400 genes coding for RBR proteins have been identified in the genomes of uni- and multicellular eukaryotes and some of their viruses, but the family has not been found in archaea or bacteria. The RBR proteins are classified into 15 major subfamilies (besides some orphan cases) by the phylogenetic relationships of the RBR segments and the conservation of their sequence architecture. The RBR domain mediates protein-protein interactions and a subset of RBR proteins has been shown to function as E3 ubiquitin ligases. RBR proteins have attracted interest because of their involvement in diseases such as parkinsonism, dementia with Lewy bodies, and Alzheimer's disease, and in susceptibility to some intracellular bacterial pathogens. Here, we present an overview of the RBR-domain containing proteins and their subcellular localization, additional domains, function, specificity, and regulation.     

Arabidopsis E2FA stimulates proliferation and endocycle separately through RBR-bound and RBR-free complexes

Post-embryonic growth in plants depends on the continuous supply of undifferentiated cells within meristems. Proliferating cells maintain their competence for division by active repression of differentiation and the associated endocycle entry. We show by upregulation and downregulation of E2FA that it is required for maintaining proliferation, as well as for endocycle entry. While E2FB-RBR1 (retinoblastoma-related protein 1) complexes are reduced after sucrose addition or at elevated CYCD3;1 levels, E2FA maintains a stable complex with RBR1 in proliferating cells. Chromatin immunoprecipitation shows that RBR1 binds in the proximity of E2F promoter elements in CCS52A1 and CSS52A2 genes, central regulators for the switch from proliferation to endocycles. Overexpression of a truncated E2FA mutant (E2FA(ΔRB)) lacking the RBR1-binding domain interferes with RBR1 recruitment to promoters through E2FA, leading to decreased meristem size in roots, premature cell expansion and hyperactivated endocycle in leaves. E2F target genes, including CCS52A1 and CCS52A2, are upregulated in E2FA(ΔRB) and e2fa knockout lines. These data suggest that E2FA in complex with RBR1 forms a repressor complex in proliferating cells to inhibit premature differentiation and endocycle entry. Thus, E2FA regulates organ growth via two distinct, sequentially operating pathways.     

Model structure of a fragment of biological knowledge (cell motility)

The aim of the study is to contribute to a better understanding of some aspects of the structure of biological knowledge and to make clearer to what extent the methods of reasoning may be useful in this field when only qualitative information is available. A fragment of biological knowledge (theory of cell motility) is analysed from the logico-methodological point of view as a coherent system and the possibility of its formal representation is investigated. The analysis is based on distinguishing the main objects and their features (attributes) of which a given piece of knowledge is composed and on the values which these features may display. The features are interconnected by relations (in which various number of arguments appear) and these relations constitute the main (general, higher level) laws of a given fragment of knowledge (theory). Values of attributes are also mutually connected and these relations correspond to the detailed (lower level) laws. A computer system (in which Prolog language was used) enables to perform inference operations of progressive as well as regressive type. The main categories of reasoning procedures are described and illustrated by examples, namely a) search for conclusions which may be confronted with the actual knowledge in order to verify the system as a whole, b) formation of working hypotheses in the process of their empirical verification and explanation of facts and laws. The problem of development and modification of the system is also discussed.     

Comparative genomics and protein domain graph analyses link ubiquitination and RNA metabolism

The human gene parkin, known to cause familial Parkinson disease, as well as several other genes, likely involved in other neurodegenerative diseases or in cancer, encode proteins of the RBR family of ubiquitin ligases. Here, we describe the structural diversity of the RBR family in order to infer their functional roles. Of particular interest is a relationship detected between RBR-mediated ubiquitination and RNA metabolism: a few RBR proteins contain RNA binding domains and DEAH-box RNA helicase domains. Global protein domain graph analyses demonstrate that this connection is not RBR-specific, but instead many other proteins contain both ubiquitination and RNA-related domains. These proteins are present in animals, plants and fungi, suggesting that the link between these two cellular processes is ancient. Our results show that global bioinformatic approaches, involving comparative genomics and domain network analyses, may unearth novel functional relationships involving well-known and thoroughly studied groups of proteins.     

Principal component analysis in combination with case-based reasoning for detecting therapeutically correct and incorrect measurements in continuous glucose monitoring systems

This paper introduces a data-driven methodology for detecting therapeutically correct and incorrect measurements in continuous glucose monitoring systems (CGMSs) in an intensive care unit (ICU). The data collected from 22 patients in an ICU with insulin therapy were obtained following the protocol established in the ICU. Measurements were classified using principal component analysis (PCA) in combination with case-based reasoning (CBR), where a PCA model was built to extract features that were used as inputs of the CBR system. CBR was trained to recognize patterns and classify these data. Experimental results showed that this methodology is a potential tool to distinguish between therapeutically correct and incorrect measurements from a CGMS, using the information provided by the monitor itself, and incorporating variables about the patient's clinical condition.     

RBR Ubiquitin Ligases: Diversification and Streamlining in Animal Lineages

The patterns of emergence and disappearance in animal species of genes encoding RBR ubiquitin ligases are described. RBR genes can be classified into subfamilies (Parkin, Ariadne, Dorfin, ARA54, etc.) according to sequence and structural data. Here, I show that most animal-specific RBR subfamilies emerged early in animal evolution, and that ancient animals, before the cnidarian/bilaterian split, had a set of RBR genes, which was as complex as the one currently found in mammals. However, some lineages (nematodes, dipteran insects) have recently suffered multiple losses, leading to a highly simplified set of RBR genes. Genes of a particular RBR subfamily, characterized by containing a helicase domain and so far found only in plants, are present also in some animal species. The meaning of these patterns of diversification and streamlining are discussed at the light of functional data. Extreme evolutionary conservation may be related to gene products having housekeeping functions. Electronic supplementary material  The online version of this article (doi:) contains supplementary material, which is available to authorized users.     

A scheduling problem for a novel container transport system: a case of mobile harbor operation schedule

Mobile Harbor (MH) is a movable floating platform with a container handling system on board so that it can load/discharge containers to/from an anchored container ship in the open sea. As with typical quay crane operation, an efficient schedule for its operation is a key to enhancing its operational productivity. A MH operation scheduling problem is to determine a timed sequence of loading/discharging tasks, assignment of MH units to each task, and their docking position, with an objective of minimizing the makespan of a series of incoming container ships. A mixed integer programming model is formulated to formally define the problem. As a practical solution method to the problem, this paper proposes a rule-based algorithm and a random key based genetic algorithm (rkGA). Computational results show that the rkGA method produces a better-quality solution than the rule-based method, while requiring longer computation time.     

Molecular insights into RBR E3 ligase ubiquitin transfer mechanisms

RING‐in‐between‐RING (RBR) ubiquitin (Ub) ligases are a distinct class of E3s, defined by a RING1 domain that binds E2 Ub‐conjugating enzyme and a RING2 domain that contains an active site cysteine similar to HECT‐type E3s. Proposed to function as RING/HECT hybrids, details regarding the Ub transfer mechanism used by RBRs have yet to be defined. When paired with RING‐type E3s, E2s perform the final step of Ub ligation to a substrate. In contrast, when paired with RBR E3s, E2s must transfer Ub onto the E3 to generate a E3~Ub intermediate. We show that RBRs utilize two strategies to ensure transfer of Ub from the E2 onto the E3 active site. First, RING1 domains of HHARI and RNF144 promote open E2~Ubs. Second, we identify a Ub‐binding site on HHARI RING2 important for its recruitment to RING1‐bound E2~Ub. Mutations that ablate Ub binding to HHARI RING2 also decrease RBR ligase activity, consistent with RING2 recruitment being a critical step for the RBR Ub transfer mechanism. Finally, we demonstrate that the mechanism defined here is utilized by a variety of RBRs.     

DQB: A novel dynamic quantitive classification model using artificial bee colony algorithm with application on gene expression profiles

In the medical domain, it is very significant to develop a rule-based classification model. This is because it has the ability to produce a comprehensible and understandable model that accounts for the predictions. Moreover, it is desirable to know not only the classification decisions but also what leads to these decisions. In this paper, we propose a novel dynamic quantitative rule-based classification model, namely DQB, which integrates quantitative association rule mining and the Artificial Bee Colony (ABC) algorithm to provide users with more convenience in terms of understandability and interpretability via an accurate class quantitative association rule-based classifier model. As far as we know, this is the first attempt to apply the ABC algorithm in mining for quantitative rule-based classifier models. In addition, this is the first attempt to use quantitative rule-based classification models for classifying microarray gene expression profiles. Also, in this research we developed a new dynamic local search strategy named DLS, which is improved the local search for artificial bee colony (ABC) algorithm. The performance of the proposed model has been compared with well-known quantitative-based classification methods and bio-inspired meta-heuristic classification algorithms, using six gene expression profiles for binary and multi-class cancer datasets. From the results, it can be concludes that a considerable increase in classification accuracy is obtained for the DQB when compared to other available algorithms in the literature, and it is able to provide an interpretable model for biologists. This confirms the significance of the proposed algorithm in the constructing a classifier rule-based model, and accordingly proofs that these rules obtain a highly qualified and meaningful knowledge extracted from the training set, where all subset of quantitive rules report close to 100% classification accuracy with a minimum number of genes. It is remarkable that apparently (to the best of our knowledge) several new genes were discovered that have not been seen in any past studies. For the applicability demand, based on the results acqured from microarray gene expression analysis, we can conclude that DQB can be adopted in a different real world applications with some modifications.     

The Balance-Scale Task Revisited: A Comparison of Statistical Models for Rule-Based and Information-Integration Theories of Proportional Reasoning

We propose and test three statistical models for the analysis of children’s responses to the balance scale task, a seminal task to study proportional reasoning. We use a latent class modelling approach to formulate a rule-based latent class model (RB LCM) following from a rule-based perspective on proportional reasoning and a new statistical model, the Weighted Sum Model, following from an information-integration approach. Moreover, a hybrid LCM using item covariates is proposed, combining aspects of both a rule-based and information-integration perspective. These models are applied to two different datasets, a standard paper-and-pencil test dataset (N = 779), and a dataset collected within an online learning environment that included direct feedback, time-pressure, and a reward system (N = 808). For the paper-and-pencil dataset the RB LCM resulted in the best fit, whereas for the online dataset the hybrid LCM provided the best fit. The standard paper-and-pencil dataset yielded more evidence for distinct solution rules than the online data set in which quantitative item characteristics are more prominent in determining responses. These results shed new light on the discussion on sequential rule-based and information-integration perspectives of cognitive development.     

Machine reasoning about anomalous sensor data

We describe a semantic data validation tool that is capable of observing incoming real-time sensor data and performing reasoning against a set of rules specific to the scientific domain to which the data belongs. Our software solution can produce a variety of different outcomes when a data anomaly or unexpected event is detected, ranging from simple flagging of data points, to data augmentation, to validation of proposed hypotheses that could explain the phenomenon. Hosted on the Jena Semantic Web Framework, the tool is completely domain-agnostic and is made domain-aware by reference to an ontology and Knowledge Base (KB) that together describe the key resources of the system being observed. The KB comprises ontologies for the sensor packages and for the domain; historical data from the network; concepts designed to guide discovery of internet resources unavailable in the local KB but relevant to reasoning about the anomaly; and a set of rules that represent domain expert knowledge of constraints on data from different kinds of instruments as well as rules that relate types of ecosystem events to properties of the ecosystem. We describe an instance of such a system that includes a sensor ontology, some rules describing coastal storm events and their consequences, and how we relate local data to external resources. We describe in some detail how a specific actual event—an unusually high chlorophyll reading—can be deduced by machine reasoning to be consistent with being caused by benthic diatom resuspension, consistent with being caused by an algal bloom, or both.     

Comparative genomics of the RBR family,including the Parkinson's disease-related gene parkin and the genes of the ariadne subfamily

Genes of the RBR family are characterized by the RBR signature (two RING finger domains separated by an IBR/DRIL domain). The RBR family is widespread in eukaryotes, with numerous members in animals (mammals, Drosophila, Caenorhabditis) and plants (Arabidopsis). But yeasts, such as Saccharomyces cerevisiae or Schizosaccharomyces pombe, contain only two RBR genes. We determined the phylogenetic relationships and the most likely orthologs in different species of several family members for which functional data are available. These include: (1) parkin, whose mutations are involved in forms of familial Parkinson's disease; (2) the ariadne genes, recently characterized in Drosophila and mammals; (3) XYbp and Dorfin, two mammalian genes whose products interact with the centrosome; (4) XAP3, RBCK1, and UIP28, mammalian genes encoding Protein Kinase-C-binding proteins; and (5) ARA54, an androgen receptor coactivator. Because several of these genes are involved in ubiquitination, we used phylogenetic and structural analyses to explore the hypothesis that all RBR proteins might play a role in ubiquitination. We show that the involvement of RBR proteins in ubiquitination predates the animals-plants-fungi divergence. On the basis of the evidence provided by cases of gene fusion, we suggest that Ariadne proteins interact with cullin domain-containing proteins to form complexes with ubiquitin-ligase activity.     

Parkin mitochondrial translocation is achieved through a novel catalytic activity coupled mechanism

Pink1, a mitochondrial kinase, and Parkin, an E3 ubiquitin ligase, function in mitochondrial maintenance. Pink1 accumulates on depolarized mitochondria, where it recruits Parkin to mainly induce K63-linked chain ubiquitination of outer membrane proteins and eventually mitophagy. Parkin belongs to the RBR E3 ligase family. Recently, it has been proposed that the RBR domain transfers ubiquitin to targets via a cysteine∼ubiquitin enzyme intermediate, in a manner similar to HECT domain E3 ligases. However, direct evidence for a ubiquitin transfer mechanism and its importance for Parkin''s in vivo function is still missing. Here, we report that Parkin E3 activity relies on cysteine-mediated ubiquitin transfer during mitophagy. Mutating the putative catalytic cysteine to serine (Parkin C431S) traps ubiquitin, and surprisingly, also abrogates Parkin mitochondrial translocation, indicating that E3 activity is essential for Parkin translocation. We found that Parkin can bind to K63-linked ubiquitin chains, and that targeting K63-mimicking ubiquitin chains to mitochondria restores Parkin C431S localization. We propose that Parkin translocation is achieved through a novel catalytic activity coupled mechanism.     

Electrolyte changes in plasma and urine of athletes during acute and rigorous bed rest and ambulatory conditions

Rigorous bed rest (RBR) induces significant electrolyte changes, but little it is not known about the effect of acute bed rest (ABR) (i.e., abrupt confinement to a RBR). The aim of this study was to measure urinary and plasma electrolyte changes during ABR and RBR conditions. The studies were done during 3 d of a pre-bed-rest (BR) period and during 7 d of an ABR and RBR period. Thirty male trained athletes aged, 24.4 ± 6.6 yr were chosen as subjects. They were divided equally into three groups: unrestricted ambulatory control subjects (UACS), acute-bed-rested subjects (ABRS), and rigorous-bed-rested subjects (RBRS). The UACS group experienced no changes in professional training and daily activities. The ABRS were submitted abruptly to a RBR regimen and without having any prior knowledge of the exact date and time when they would be subjected to an RBR regimen. The RBRS were subjected to an RBR regime on a predetermined date and time known to them from the beginning of the study. Sodium (Na), potassium (K), magnesium (Mg), calcium (Ca), and phosphate (P) in plasma and urine, plasma renin activity (PRA) and plasma aldosterone (PA), physical characteristics, peak oxygen uptake, and food and water intakes were measured. Urinary Na, K, Ca, Mg, and P excretion and plasma Na, K, Mg, Ca, and P concentration, PRA, and PA concentration increased significantly (p ≤ 0.01), whereas body weight, peak oxygen uptake, and food and water intakes decreased significantly in the ABRS and RBRS groups when compared with the UACS group. However, urinary and plasma Na, K, Mg, P, and Ca, PRA, and PA values increased much faster and were much greater in the ABRS group than in the RBRS group. Plasma and urinary Na, K, Ca, Mg, and P, PRA and PA levels, food and water intakes, body weight, and peak oxygen uptake did not change significantly in the UACS group when compared with its baseline control values. It was shown that RBR and ABR conditions induce significant increases in urinary and plasma electrolytes; however, urinary and plasma electrolyte changes appeared much faster and were much greater in the ABRS group than the RBRS group. It was concluded that the more abruptly motor activity is ended, the faster and the greater the urinary and plasma electrolyte change.