Conditional Probability-Based Significance Tests for Sequential Patterns in Multineuronal Spike Trains

We consider the problem of detecting statistically significant sequential patterns in multineuronal spike trains. These patterns are characterized by ordered sequences of spikes from different neurons with specific delays between spikes. We have previously proposed a data-mining scheme to efficiently discover such patterns, which occur often enough in the data. Here we propose a method to determine the statistical significance of such repeating patterns. The novelty of our approach is that we use a compound null hypothesis that not only includes models of independent neurons but also models where neurons have weak dependencies. The strength of interaction among the neurons is represented in terms of certain pair-wise conditional probabilities. We specify our null hypothesis by putting an upper bound on all such conditional probabilities. We construct a probabilistic model that captures the counting process and use this to derive a test of significance for rejecting such a compound null hypothesis. The structure of our null hypothesis also allows us to rank-order different significant patterns. We illustrate the effectiveness of our approach using spike trains generated with a simulator.

Efficient content-based image retrieval based on multi-feature fusion

The increased availability of image capturing devices has enabled collections of digital images to rapidly expand in both size and diversity. This has created a constantly growing need for efficient and effective image browsing, searching, and retrieval tools. Pseudo-relevance feedback (PRF) has proven to be an effective mechanism for improving retrieval accuracy. An original, simple yet effective rank-based PRF mechanism (RB-PRF) that takes into account the initial rank order of each image to improve retrieval accuracy is proposed. This RB-PRF mechanism innovates by making use of binary image signatures to improve retrieval precision by promoting images similar to highly ranked images and demoting images similar to lower ranked images. Empirical evaluations based on standard benchmarks, namely Wang, Oliva & Torralba, and Corel datasets demonstrate the effectiveness of the proposed RB-PRF mechanism in image retrieval.

Epidemiology of candidemia in Finland

Candida species are an important cause of nosocomial bloodstream infections in hospitalized patients worldwide, with associated high mortality, excess length of stay and costs. Main contributors to candidemias is profound immunosuppression due to serious underlying condition or intensive treatments leading to an increasing number of susceptible patients. The rank order of causative Candida species varies over time and in different geographic locations. The aim of this study was to obtain information on epidemiology of candidemia in Finland, to identify trends in incidence, causative species, and patient populations at risk. In order to reveal possible outbreaks and assess the value of one molecular typing method, restriction enzyme analysis (REA), in epidemiological study, we analyzed C. albicans bloodstream isolates in Uusimaa region in Southern Finland during eight years. The data from the National Infectious Disease Register were used to assess the incidence and epidemiological features of candidemia cases. In Helsinki University Central Hospital (HUCH) all patients with blood culture yielding any Candida spp. were identified from laboratory log-books and from Finnish Hospital Infection Program. All the patients with a stored blood culture isolate of C. albicans were identified through microbiology laboratory logbooks, and stored isolates were genotyped with REA in the National Institute for Health and Welfare (former KTL). The incidence of candidemia in Finland is globally relatively low, but increased between between 1990s and 2000s. The incidence was highest in males >65 years of age, but incidence rates for patients <1-15 years were lower during 2000s than during 1990s. In HUCH the incidence of candidemia remained low and constant during our 18 years of observation, but a significant shift in patient-populations at risk was observed, associated with patients treated in intensive care units, such as premature neonates and surgical patients. The predominating causative species in Finland and in HUCH is C. albicans, but the proportion of C. glabrata increased considerably. The crude one-month case fatality was constantly high between 28-33%. REA differentiated efficiently between C. albicans blood culture isolates and no clusters were observed in the hospitals involved, despite of abundant transfer of patients among them. Candida spp. are an important cause of nosocomial blood stream infections in Finland, and continued surveillance is necessary to determine the overall trends and patient groups at risk, and reduce the impact of these infections in the future. Molecular methods provide an efficient tool for investigation of suspected outbreak and should be available in the future in Finland, also.

Design of speaker identification schemes for large number of speakers (A)

Design of speaker identification schemes for a small number of speakers (around 10) with a high degree of accuracy in controlled environment is a practical proposition today. When the number of speakers is large (say 50–100), many of these schemes cannot be directly extended, as both recognition error and computation time increase monotonically with population size. The feature selection problem is also complex for such schemes. Though there were earlier attempts to rank order features based on statistical distance measures, it has been observed only recently that the best two independent measurements are not the same as the combination in two's for pattern classification. We propose here a systematic approach to the problem using the decision tree or hierarchical classifier with the following objectives: (1) Design of optimal policy at each node of the tree given the tree structure i.e., the tree skeleton and the features to be used at each node. (2) Determination of the optimal feature measurement and decision policy given only the tree skeleton. Applicability of optimization procedures such as dynamic programming in the design of such trees is studied. The experimental results deal with the design of a 50 speaker identification scheme based on this approach.

Impact of depressive symptoms on subjective well-being: the importance of patient-reported outcomes in schizophrenia

Objective: The subjective experience of psychotic patients toward treatment is a key factor in medication adherence, quality of life, and clinical outcome. The aim of this study was to assess the subjective well-being in patients with schizophrenia and to examine its relationship with the presence and severity of depressive symptoms. Methods: A multicenter, cross-sectional study was conducted with clinically stable outpatients diagnosed with schizophrenia. The Subjective Well-Being under Neuroleptic Scale - short version (SWN-K) and the Calgary Depression Scale for Schizophrenia (CDSS) were used to gather information on well-being and the presence and severity of depressive symptoms, respectively. Spearman's rank correlation was used to assess the associations between the SWN-K total score, its five subscales, and the CDSS total score. Discriminative validity was evaluated against that criterion by analysing the area under the curve (AUC). Results: Ninety-seven patients were included in the study. Mean age was 35 years (standard deviation = 10) and 72% were male. Both the total SWN-K scale and its five subscales correlated inversely and significantly with the CDSS total score (P < 0.0001). The highest correlation was observed for the total SWN-K (Spearman's rank order correlation [ rho] = -0.59), being the other correlations: mental functioning (-0.47), social integration (-0.46), emotional regulation (-0.51), physical functioning (-0.48), and self-control (-0.41). A total of 33 patients (34%) were classified as depressed. Total SWN-K showed the highest AUC when discriminating between depressive severity levels (0.84), followed by emotional regulation (0.80), social integration (0.78), physical functioning and self-control (0.77), and mental functioning (0.73). Total SWN-K and its five subscales showed a significant linear trend against CDSS severity levels (P < 0.001). Conclusion: The presence of moderate to severe depressive symptoms was relatively high, and correlated inversely with patients' subjective well-being. Routine assessment of patient-reported measures in patients with schizophrenia might reduce potential discrepancy between patient and physician assessment, increase therapeutic alliance, and improve outcome.

Intersexual and spatial heterogeneity in trophic attributes of the sleeper, Bostrychus africanus (Eleotridae) in the Qua Iboe Estuary, Nigeria

There was variation in the ingestion of the food objects by the sexes. Despite the similarity in rank-order of the food objects, the ingestion of the objects vary significantly (rg=0.320, P>0.05). Dipterans adult and Hymenoptera were the only food objects not eaten by the males whereas insect remains and unidentified bivalves were absent from the trophics spectrum of the females. There was significant increase in feeding intensity by females than males. There was significant increase in GRI by specimens from Nipa Creek whereas individuals from mangrove creek recorded higher MGF and vice-versa. Dipterans adult. Hymenoptera, insect remains, Neritina glabrata and unid bivalves were absent from dietaries for nipa creek whereas a complete array of the food objects were eaten in the mangrove creek. The present findings highlights the importance of the mangrove ecosystem as the native vegetation encompassing great diversity of food resources and living conditions than the succeeding alien nipa vegetation

Fuzzy inferring mathematical morphology and optical implementation

Fuzzification is introduced into gray-scale mathematical morphology by using two-input one-output fuzzy rule-based inference systems. The fuzzy inferring dilation or erosion is defined from the approximate reasoning of the two consequences of a dilation or an erosion and an extended rank-order operation. The fuzzy inference systems with numbers of rules and fuzzy membership functions are further reduced to a simple fuzzy system formulated by only an exponential two-input one-output function. Such a one-function fuzzy inference system is able to approach complex fuzzy inference systems by using two specified parameters within it-a proportion to characterize the fuzzy degree and an exponent to depict the nonlinearity in the inferring. The proposed fuzzy inferring morphological operators tend to keep the object details comparable to the structuring element and to smooth the conventional morphological operations. Based on digital area coding of a gray-scale image, incoherently optical correlation for neighboring connection, and optical thresholding for rank-order operations, a fuzzy inference system can be realized optically in parallel. (C) 1996 Society of Photo-Optical Instrumentation Engineers.

Responses to seasonal changes in nutrient quality and patchiness of food in a multigroup community of Tibetan macaques at Mt. Emei

Provisioning along pedestrian trails by tourists much increased the nutrient quality and patchiness of food (NqPF)for Tibetan macaques (Macaca thibetana) at Mt Emei in spring and summer. In the habitat at a temperate-subtropical transition zone, the mncaque's NqPF could be ordered in a decreasing rank from spring summer to autumn to winter With the aid of a radio-tracking system, I collected ranging data on a multigroup community in three 70-day periods representing the different seasons in 1991-92, Rank-order correlation on the data show that with the decline of NqPF; the groups tended to increase days away from the trail, their effective range size (ERS) their exclusive area (EA) and the number of days spent in the EA, and reduced their group/community density and the ratio of the overlapped range to the seasonal range (ROR). In icy/snowy winter; the macaques searched for mature leaves slowly and carefully in the largest seasonal range with a considerable portion that was nor used in other seasons. Of the responses, the ROR decreased with the reduction in group/community density; and the ERS was the function of both group size (+) and intergroup rank (-) when favorite food was highly clumped. All above responses were clearly bound to maximize foraging effectiveness and minimize energy expenditure, and their integration in term of changes in time and space leads to better understanding macaque ecological adaptability. Based on this study and previous work on behavioral and physiological factors, I suggest a unifying theory of intergroup interactions. Ir! addition, as the rate of behavioral interactions,was also related to the group density, I Waser's (1976) gas model probably applies to behavioral, as well as spatial, data on intergroup interactions.

Variability in north temperate dung beetle assemblages at different spatial and temporal scales

This study examined the spatial and temporal variability of dung beetle assemblages across a variety of scales e.g. from the between-pad scale (examining the effects of dung size and type) to larger spatial scales encompassing southern Ireland. Dung beetle assemblage structure as sampled by dung pad cohort samples and dung baited pitfall trapping were compared. Generally, the rank order of abundance of dung beetle species was significantly correlated between pitfall catches and cohort pad samples. Across different dung sizes, in both pitfall catches and cohort pad samples, the relative abundance of species was frequently significantly different, but the rank order of abundance of dung beetle was usually significantly correlated. Considerable variations in pitfall catches at temporal scales of a few days appeared to be closely related to weather conditions and rotational grazing. However, despite considerable variation in absolute abundances between consecutive days of sampling, assemblage structure typically remained very similar. The relationship between dung pad size and dung beetle colonisation was investigated. In field experiments in which pads of different sizes (0.25 L, 0.5 L, 1.0 L and 1.5 L) were artificially deposited, there was a positive relationship between pad size and both biomass and number of beetles colonising dung pads and pitfall traps. In addition, with one exception, the field experiments indicated a general positive relationship between dung pad size and biomass density (dung beetle biomass per unit dung volume). A laboratory experiment indicated that pat residence times of A. rufipes were significantly correlated with dung pad size. Investigation of naturally-deposited cow dung pads in the field also indicated that both larval numbers and densities were significantly correlated with dung pad size. These results were discussed in the context of theory related to aggregation and coexistence of species, and resource utilisation by organisms in ephemeral, patchy resources. The colonisation by dung beetles of dung types from native herbivores (sheep, horse and cow) was investigated in field experiments. There were significant differences between the dung types in the chemical parameters measured, and there were significant differences in abundances of dung beetles colonising the dung types. Sheep dung was typically the preferred dung type. Data from these field experiments, and from published literature, indicated that dung beetle species can display dung type preferences, in terms of comparisons of both absolute and relative abundances. In addition, data from laboratory experiments indicate that both Aphodius larval production and pat residence times tended to be higher in those dung types which were preferred by adult Aphodius in the colonisation experiments. Data from dung-baited pitfall trapping (from this and another study) at several sites (up to 180 km distant) and over a number of years (between 1991 and 1996) were used to investigate spatial and temporal variation in dung beetle assemblage structure and composition (Aphodius, Sphaeridium and Geotrupes) across a range of scales in southern Ireland. Species richness levels, species composition and rank order of abundances were very similar between the assemblages. The temporal variability between seasons within any year exceeded temporal variability between years. DCA ordinations indicated that there was a similar level of variability between assemblage structure from the between-field (~1km) to regional (~180 km) spatial scales, and between year (6 years) temporal scales. At the biogeographical spatial scale, analysis of data from the literature indicated that there was considerable variability at this scale, largely due to species turnover.

In vitro-in vivo correlations of self-emulsifying drug delivery systems combining the dynamic lipolysis model and neuro-fuzzy networks

The aim of the current study was to evaluate the potential of the dynamic lipolysis model to simulate the absorption of a poorly soluble model drug compound, probucol, from three lipid-based formulations and to predict the in vitro-in vivo correlation (IVIVC) using neuro-fuzzy networks. An oil solution and two self-micro and nano-emulsifying drug delivery systems were tested in the lipolysis model. The release of probucol to the aqueous (micellar) phase was monitored during the progress of lipolysis. These release profiles compared with plasma profiles obtained in a previous bioavailability study conducted in mini-pigs at the same conditions. The release rate and extent of release from the oil formulation were found to be significantly lower than from SMEDDS and SNEDDS. The rank order of probucol released (SMEDDS approximately SNEDDS > oil formulation) was similar to the rank order of bioavailability from the in vivo study. The employed neuro-fuzzy model (AFM-IVIVC) achieved significantly high prediction ability for different data formations (correlation greater than 0.91 and prediction error close to zero), without employing complex configurations. These preliminary results suggest that the dynamic lipolysis model combined with the AFM-IVIVC can be a useful tool in the prediction of the in vivo behavior of lipid-based formulations.

Ranking of length-class seasonal and regional effects on dietary compositions of the co-occurring Pagrus auratus (Sparidae) and Pseudocaranx georgianus (Carangidae)

Using an effective combination of multivariate testing and ordination analyses, this study compares the extents to which the diets of two co-occurring fish species (Pagrus auratus and Pseudocaranx georgianus) are related to body size (length class), season and region and the rank order importance of those effects. Thus, volumetric dietary compositions were determined for these species on the lower west coast of Australia, where both are abundant, and for P. auratus from the mid west coast and P. georgianus from the south coast. The diet of P. auratus on the lower west coast was strongly related to body size and slightly less to season. With increasing body size, its diet shifted from predominantly ophiuroids to larger prey, such as brachyuran crabs, teleosts, echinoids and ultimately asteroids, probably reflecting a shift from foraging over soft sediments to areas over and around reefs. Seasonal changes on the lower west coast were restricted mainly to small P. auratus, while larger fish underwent seasonal changes further north. Analyses using a common size range of medium to larger P. auratus demonstrated that dietary composition differed more between regions than seasons. The relationships between diet and length class of P. georgianus on both the lower west and south coasts were less pronounced than for P. auratus and seasonal changes were restricted to the south coast, where amphipod consumption increased markedly in summer. The diet of P. georgianus was related far more to region than length class and season, with more small teleosts, small crabs, carideans and littorinids and less amphipods, isopods and small bivalves being ingested on the lower west than south coasts. Although crabs and teleosts were important typifying prey of P. auratus and P. georgianus, when co-occurring, the former predator tended to ingest greater volumes of larger and often less mobile prey. This reflects differences in dentition, jaw morphology and feeding behaviour and reduces the potential for competition for food resources. The results imply that P. auratus and P. georgianus are opportunistic feeders and that the effects of length class, season and region on dietary composition and their rank orders can vary markedly between species and for length class and season between regions for the same species.

Synthesis of Peptidyl Ene Diones: Selective Inactivators of the Cysteine Proteinases

A series of synthetic peptides in which the C-terminal carboxyl grouping (-CO2H) of each has been chemically converted into a variety of ene dione derivatives (-CO-CH CH-CO-X; X -H, -Me, -OBut, - OEt, -OMe, -CO-OMe), have been prepared and tested as inactivators against typical members of the serine and cysteine protease families. For example, the sequences Cbz-Pro-Phe-CH CH-CO-OEt (I) which fulfils the known primary and secondary specificity requirements of the serine protease chymotrypsin, and Cbz-Phe-Ala-CH CH-CO-OEt (II) which represents a general recognition sequence for cysteine proteases such as cathepsins B, L and S, have been tested as putative irreversible inactivators of their respective target proteases. It was found that, whereas II, for example, functioned as a time-dependent, irreversible inactivator of each of the cysteine proteases, I behaved only as a modest competitive reversible inhibitor of chymotrypsin. Within the simple ester sequences Cbz- Phe-Ala-CH CH-CO-R, the rank order of inhibitor effectiveness decreases in the order R -OMe > - OEt >> -OBut. It was also found that the presence of both an unsaturated double bond and an ester (or a-keto ester) moiety were indispensable for obtaining irreversible inactivators. Of the irreversible inactivators synthesized, Cbz-Phe-Ala-CH CHCO- CO-OEt (which contains a highly electrophilic a-keto ester grouping) was found to be the most effective exhibiting, for example, second-order rate constants of approximately 1.7 106/M/min and approximately 4.9 104/M/min against recombinant human cathepsin S and human spleenic cathepsin B, respectively. This initial study thus holds out the promise that this class of inactivator may well be specific for the cysteine protease subclass.

Extending Zipf’s law to n-grams for large corpora

Experiments show that for a large corpus, Zipf’s law does not hold for all rank of words: the frequencies fall below those predicted by Zipf’s law for ranks greater than about 5,000 word types in the English language and about 30,000 word types in the inflected languages Irish and Latin. It also does not hold for syllables or words in the syllable-based languages, Chinese or Vietnamese. However, when single words are combined together with word n-grams in one list and put in rank order, the frequency of tokens in the combined list extends Zipf’s law with a slope close to -1 on a log-log plot in all five languages. Further experiments have demonstrated the validity of this extension of Zipf’s law to n-grams of letters, phonemes or binary bits in English. It is shown theoretically that probability theory
alone can predict this behavior in randomly created n-grams of binary bits.

Identification of high-affinity binding sites for the insulin sensitizer rosiglitazone (BRL-49653) in rodent and human adipocytes using a radioiodinated ligand for peroxisomal proliferator-activated receptor gamma.

A radioiodinated ligand, [125I]SB-236636 [(S)-(-)3-[4-[2-[N-(2-benzoxazolyl)-N-methylamino]ethoxy]3-[125I]iodophenyl]2-ethoxy propanoic acid], which is specific for the ? isoform of the peroxisomal proliferator activated receptor (PPAR?), was developed. [125I]SB-236636 binds with high affinity to full-length human recombinant PPAR?1 and to a GST (glutathione S-transferase) fusion protein contg. the ligand binding domain of human PPAR?1 (KD = 70 nM). Using this ligand, the authors characterized binding sites in adipose-derived cells from rat, mouse and humans. In competition expts., rosiglitazone (BRL-49653), a potent antihyperglycemic agent, binds with high affinity to sites in intact adipocytes (IC50 = 12, 4 and 9 nM for rat, 3T3-L1 and human adipocytes, resp.). Binding affinities (IC50) of other thiazolidinediones for the ligand binding domain of PPAR?1 were comparable with those detd. in adipocytes and reflected the rank order of potencies of these agents as stimulants of glucose transport in 3T3-L1 adipocytes and antihyperglycemic agents in vivo: rosiglitazone > pioglitazone > troglitazone. Competition of [125I]SB-236636 binding was stereoselective in that the IC50 value of SB-219994, the (S)-enantiomer of an ?-trifluoroethoxy propanoic acid insulin sensitizer, was 770-fold lower than that of SB-219993 [(R)-enantiomer] at recombinant human PPAR?1. The higher binding affinity of SB-219994 also was evident in intact adipocytes and reflected its 100-fold greater potency as an antidiabetic agent. The results strongly suggest that the high-affinity binding site for [125I]SB-236636 in intact adipocytes is PPAR? and that the pharmacol. of insulin-sensitizer binding in rodent and human adipocytes is very similar and, moreover, predictive of antihyperglycemic activity in vivo.

Induction and rejoining of DNA double-strand breaks in bladder tumor cells

The induction and rejoining of radiation-induced double-strand breaks (DSBs) in cells of six bladder tumor cell lines (T24, UM-UC3, TCC-SUP, RT112, J82, HT1376) were measured using the neutral comet assay. Radiation dose-response curves (0-60 Gy) showed damage (measured as mean tail moment) for five of the cell lines in the same rank order as cell survival (measured over 0-10 Gy), with the least damage in the most radioresistant cell line. Damage induction correlated well with clonogenic survival at high doses (SF10) for all six cell lines. At the clinically relevant dose of 2 Gy, correlation was good for four cell lines but poor for two (TCC-SUP and T24), The rejoining process had a fast and slow component for all cell lines. The rate of these two components of DNA repair did not correlate with cell survival. However, the time taken to reduce the amount of DNA damage to preirradiated control levels correlated positively with cell survival at 10 Gy but not 2 Gy; radioresistant cells rejoined the induced DSBs to preirradiation control levels more quickly than the radiosensitive cells. Although the results show good correlation between SF10 and DSBs for all six cell lines, the lack of correlation with SF2 for TCC-SUP and T24 cells would suggest that a predictive test should be carried out at the clinically relevant dose. At present the neutral comet assay cannot achieve this. (C) 2000 by Radiation Research Society.