Efficient learning of decomposable models with a bounded clique size

The learning of probability distributions from data is a ubiquitous problem in the fields of Statistics and Artificial Intelligence. During the last decades several learning algorithms have been proposed to learn probability distributions based on decomposable models due to their advantageous theoretical properties. Some of these algorithms can be used to search for a maximum likelihood decomposable model with a given maximum clique size, k, which controls the complexity of the model. Unfortunately, the problem of learning a maximum likelihood decomposable model given a maximum clique size is NP-hard for k > 2. In this work, we propose a family of algorithms which approximates this problem with a computational complexity of O(k · n^2 log n) in the worst case, where n is the number of implied random variables. The structures of the decomposable models that solve the maximum likelihood problem are called maximal k-order decomposable graphs. Our proposals, called fractal trees, construct a sequence of maximal i-order decomposable graphs, for i = 2, ..., k, in k − 1 steps. At each step, the algorithms follow a divide-and-conquer strategy based on the particular features of this type of structures. Additionally, we propose a prune-and-graft procedure which transforms a maximal k-order decomposable graph into another one, increasing its likelihood. We have implemented two particular fractal tree algorithms called parallel fractal tree and sequential fractal tree. These algorithms can be considered a natural extension of Chow and Liu’s algorithm, from k = 2 to arbitrary values of k. Both algorithms have been compared against other efficient approaches in artificial and real domains, and they have shown a competitive behavior to deal with the maximum likelihood problem. Due to their low computational complexity they are especially recommended to deal with high dimensional domains.

Catches per unit of effort of bigeye tuna: a new analysis with regression trees and simulated annealing

ENGLISH: We analyzed catches per unit of effort (CPUE) from the Japanese longline fishery for bigeye tuna (Thunnus obesus) in the central and eastern Pacific Ocean (EPO) with regression tree methods. Regression trees have not previously been used to estimate time series of abundance indices fronl CPUE data. The "optimally sized" tree had 139 parameters; year, month, latitude, and longitude interacted to affect bigeye CPUE. The trend in tree-based abundance indices for the EPO was similar to trends estimated from a generalized linear model and fronl an empirical model that combines oceanographic data with information on the distribution of fish relative to environmental conditions. The regression tree was more parsimonious and would be easier to implement than the other two nl0dels, but the tree provided no information about the nlechanisms that caused bigeye CPUEs to vary in time and space. Bigeye CPUEs increased sharply during the mid-1980's and were more variable at the northern and southern edges of the fishing grounds. Both of these results can be explained by changes in actual abundance and changes in catchability. Results from a regression tree that was fitted to a subset of the data indicated that, in the EPO, bigeye are about equally catchable with regular and deep longlines. This is not consistent with observations that bigeye are more abundant at depth and indicates that classification by gear type (regular or deep longline) may not provide a good measure of capture depth. Asimulated annealing algorithm was used to summarize the tree-based results by partitioning the fishing grounds into regions where trends in bigeye CPUE were similar. Simulated annealing can be useful for designing spatial strata in future sampling programs. SPANISH: Analizamos la captura por unidad de esfuerzo (CPUE) de la pesquería palangrera japonesa de atún patudo (Thunnus obesus) en el Océano Pacifico oriental (OPO) y central con métodos de árbol de regresión. Hasta ahora no se han usado árboles de regresión para estimar series de tiempo de índices de abundancia a partir de datos de CPUE. EI árbol de "tamaño optimo" tuvo 139 parámetros; ano, mes, latitud, y longitud interactuaron para afectar la CPUE de patudo. La tendencia en los índices de abundancia basados en árboles para el OPO fue similar a las tendencias estimadas con un modelo lineal generalizado y con un modelo empírico que combina datos oceanográficos con información sobre la distribución de los peces en relación con las condiciones ambientales. EI árbol de regresión fue mas parsimonioso y seria mas fácil de utilizar que los dos otros modelos, pero no proporciono información sobre los mecanismos que causaron que las CPUE de patudo valiaran en el tiempo y en el espacio. Las CPUE de patudo aumentaron notablemente a mediados de los anos 80 y fueron mas variables en los extremos norte y sur de la zona de pesca. Estos dos resultados pueden ser explicados por cambios en la abundancia real y cambios en la capturabilidad. Los resultados de un arbal de regresión ajustado a un subconjunto de los datos indican que, en el OPO, el patudo es igualmente capturable con palangres regulares y profundos. Esto no es consistente con observaciones de que el patudo abunda mas a profundidad e indica que clasificación por tipo de arte (palangre regular 0 profundo) podría no ser una buena medida de la profundidad de captura. Se uso un algoritmo de templado simulado para resumir los resultados basados en el árbol clasificando las zonas de pesca en zonas con tendencias similares en la CPUE de patudo. El templado simulado podría ser útil para diseñar estratos espaciales en programas futuros de muestreo. (PDF contains 45 pages.)

Triploidy induction in Heterobranchus longifilis (Family: Clariidae) by cold shock

Triploid was induced in African Catfish (Heterobranchus longifilis) by cold shocking activated eggs at 5 degree C for forty minutes starting 3-4 minutes after fertilization. Triploidy was confirmed from mitotic chromosomes prepared from embryo which showed 100% triploidy in the cold shocking treatment and 100% diploidy in the control treatment

Effects of some environmental factors on the fecundity of Tilapia species (Family Cichlidae) in Kigera Reservoir, New Bussa

The effects of some environmental factors on the fecundity of Tilapia species (Family Cichilidae) was carried out at the Kigera dam. Four Tilapia species caught were Tilapia zilli, Hemichromis fasciatus, Sarotherodon galilaeus and Oreochromis niloticus while the environmental factors considered were water temperature, Dissolved Oxygen, pH value, level of rainfall and rate of sunshine and range of time. 43 fish comprising of 25 male with (58.1%) and 18 females having (41.9%) were studied with 74.42% been sexually matured. Both high levels of rainfall and dissolved oxygen favoured fecundity. The spawning peak occurred in (July), environmental factors monitored indicated that dissolved oxygen ranges from 3.7 to 4.45mg/lit rainfall ranges from (34.90mm to 237.80mm) sunshine ranges from (5hrs-8hrs) and pH ranges from (7.35-7.45). The spawning of these species in their natural or hatchery condition is therefore best achieved during the peak of raining season

Expression of eph-family receptor tyrosine kinases and ephrins in the tadpole of the frog Xenopus laevis, and possible roles in the development of retinotectal topography

Assembling a nervous system requires exquisite specificity in the construction of neuronal connectivity. One method by which such specificity is implemented is the presence of chemical cues within the tissues, differentiating one region from another, and the presence of receptors for those cues on the surface of neurons and their axons that are navigating within this cellular environment.

Connections from one part of the nervous system to another often take the form of a topographic mapping. One widely studied model system that involves such a mapping is the vertebrate retinotectal projection-the set of connections between the eye and the optic tectum of the midbrain, which is the primary visual center in non-mammals and is homologous to the superior colliculus in mammals. In this projection the two-dimensional surface of the retina is mapped smoothly onto the two-dimensional surface of the tectum, such that light from neighboring points in visual space excites neighboring cells in the brain. This mapping is implemented at least in part via differential chemical cues in different regions of the tectum.

The Eph family of receptor tyrosine kinases and their cell-surface ligands, the ephrins, have been implicated in a wide variety of processes, generally involving cellular movement in response to extracellular cues. In particular, they possess expression patterns-i.e., complementary gradients of receptor in retina and ligand in tectum- and in vitro and in vivo activities and phenotypes-i.e., repulsive guidance of axons and defective mapping in mutants, respectively-consistent with the long-sought retinotectal chemical mapping cues.

The tadpole of Xenopus laevis, the South African clawed frog, is advantageous for in vivo retinotectal studies because of its transparency and manipulability. However, neither the expression patterns nor the retinotectal roles of these proteins have been well characterized in this system. We report here comprehensive descriptions in swimming stage tadpoles of the messenger RNA expression patterns of eleven known Xenopus Eph and ephrin genes, including xephrin-A3, which is novel, and xEphB2, whose expression pattern has not previously been published in detail. We also report the results of in vivo protein injection perturbation studies on Xenopus retinotectal topography, which were negative, and of in vitro axonal guidance assays, which suggest a previously unrecognized attractive activity of ephrins at low concentrations on retinal ganglion cell axons. This raises the possibility that these axons find their correct targets in part by seeking out a preferred concentration of ligands appropriate to their individual receptor expression levels, rather than by being repelled to greater or lesser degrees by the ephrins but attracted by some as-yet-unknown cue(s).

The variations and climatic significance of D/H ratios in the carbon-bound hydrogen of cellulose in trees

The σD values of nitrated cellulose from a variety of trees covering a wide geographic range have been measured. These measurements have been used to ascertain which factors are likely to cause σD variations in cellulose C-H hydrogen.

It is found that a primary source of tree σD variation is the σD variation of the environmental precipitation. Superimposed on this are isotopic variations caused by the transpiration of the leaf water incorporated by the tree. The magnitude of this transpiration effect appears to be related to relative humidity.

Within a single tree, it is found that the hydrogen isotope variations which occur for a ring sequence in one radial direction may not be exactly the same as those which occur in a different direction. Such heterogeneities appear most likely to occur in trees with asymmetric ring patterns that contain reaction wood. In the absence of reaction wood such heterogeneities do not seem to occur. Thus, hydrogen isotope analyses of tree ring sequences should be performed on trees which do not contain reaction wood.

Comparisons of tree σD variations with variations in local climate are performed on two levels: spatial and temporal. It is found that the σD values of 20 North American trees from a wide geographic range are reasonably well-correlated with the corresponding average annual temperature. The correlation is similar to that observed for a comparison of the σD values of annual precipitation of 11 North American sites with annual temperature. However, it appears that this correlation is significantly disrupted by trees which grew on poorly drained sites such as those in stagnant marshes. Therefore, site selection may be important in choosing trees for climatic interpretation of σD values, although proper sites do not seem to be uncommon.

The measurement of σD values in 5-year samples from the tree ring sequences of 13 trees from 11 North American sites reveals a variety of relationships with local climate. As it was for the spatial σD vs climate comparison, site selection is also apparently important for temporal tree σD vs climate comparisons. Again, it seems that poorly-drained sites are to be avoided. For nine trees from different "well-behaved" sites, it was found that the local climatic variable best related to the σD variations was not the same for all sites.

Two of these trees showed a strong negative correlation with the amount of local summer precipitation. Consideration of factors likely to influence the isotopic composition of summer rain suggests that rainfall intensity may be important. The higher the intensity, the lower the σD value. Such an effect might explain the negative correlation of σD vs summer precipitation amount for these two trees. A third tree also exhibited a strong correlation with summer climate, but in this instance it was a positive correlation of σD with summer temperature.

The remaining six trees exhibited the best correlation between σD values and local annual climate. However, in none of these six cases was it annual temperature that was the most important variable. In fact annual temperature commonly showed no relationship at all with tree σD values. Instead, it was found that a simple mass balance model incorporating two basic assumptions yielded parameters which produced the best relationships with tree σD values. First, it was assumed that the σD values of these six trees reflected the σD values of annual precipitation incorporated by these trees. Second, it was assumed that the σD value of the annual precipitation was a weighted average of two seasonal isotopic components: summer and winter. Mass balance equations derived from these assumptions yielded combinations of variables that commonly showed a relationship with tree σD values where none had previously been discerned.

It was found for these "well-behaved" trees that not all sample intervals in a σD vs local climate plot fell along a well-defined trend. These departures from the local σD VS climate norm were defined as "anomalous". Some of these anomalous intervals were common to trees from different locales. When such widespread commonalty of an anomalous interval occurred, it was observed that the interval corresponded to an interval in which drought had existed in the North American Great Plains.

Consequently, there appears to be a combination of both local and large scale climatic information in the σD variations of tree cellulose C-H hydrogen.

Structure of the limbs and its significance for the ecology and systematics of the family Chydoridae. [Translation of: Voprosy Gidrobiologii, p385, 1965]

For an explanation of the dynamics of numbers of chydorids, appearing a massive group in the littoral of fresh water bodies, the structure of the limbs of 29 species was studied.

Key to the larvae of the genera of the sub-family Orthocladiinae from Larvae and Pupae of midges of the sub-family Orthocladiinae of the fauna of the USSR. [Translation from: Larvae and pupae of midges of the subfamily Orthocladiinae of the fauna of the USSR (Diptera, Chironomidae = Tendipedidae) (ed. Pankratova V.Ya.), 51-55. Leningrad, Nauka, 1970. ]

A key to the larvae of the genera of the sub-family Orthocladiinae from Larvae and Pupae of midges of the sub-family Orthocladiinae. Parts of the key refer to the rest of the publication which is not included in this partial translation.

Family: Cladophora Kutzing 1843. identification key. [ Translation from: Flora Slodkowodna Polski 10, 227-263, 1972.]

A description of the algal genus Cladophora from Vol 10 of the ”Freshwater Flora of Poland”. Illustrations are included.

The family of chironomids or midges - Chironomidae. Keys to larvae and pupae [Translation from: Guide to the freshwater invertebrates of the European area of the USSR (Plankton and benthos) (ed. L. A. Kutikova and Ya. I. Starobogatov), p.371-431. Leningrad, Gidrometeoizdat, 1977]

The translation of this section of the larger publication ”Opredelitel' presnovodnykh bespozvonochnykh evropeiskoi chasti SSSR. (Plankton i bentos)” provides identification keys to the larvae and pupae of chironomids that occur in the Soviet Union. The morphology of the larvae of Chironomidae is described in the introductory part.

Relações filogenéticas dos tubarões-martelo (Elasmobranchii: Sphyrnidae) com ênfase em morfologia

A cabeça lateralmente expandida é a principal sinapomorfia da família Sphyrnidae, a qual compreende todos os tubarões-martelo. Apesar de haver trabalhos abordando esta família, sua anatomia interna tem sido negligenciada como fonte de caracteres para serem utilizados na taxonomia e filogenia dos Sphyrnidae, em especial acerca de seu cefalofólio. Além disso, outros caracteres permanecem pouco estudados na família, tais como os referentes à anatomia dentária, dentículos dérmicos e sistema de poros sensoriais do cefalofólio. As relações filogenéticas entre os Sphyrnidae, e desta família entre os demais Carcharhiniformes, ainda é controversa como sua taxonomia. As poucas hipóteses filogenéticas conhecidas, baseadas em morfologia, não foram testadas sob critérios sistemáticos. Já as poucas hipóteses moleculares existentes foram testadas mas são ainda mais controversas. O presente estudo apresenta uma revisão anatômica e taxonômica dos Sphyrnidae, construindo uma matriz de caracteres mais robusta para testar as relações filogenéticas entre os Sphyrnidae, contando com 3 gêneros relacionados no grupo externo: Carcharhinus, Rhizoprionodon e Negaprion. Os resultados posicionam S. tiburo na base da família, e E. blochii e S. tudes compondo um dos clados mais derivados. Espécies de maior porte e cefalofólio mais expandido compõem um clado formado por S. mokarran + (S. lewini + S. zygaena), ao passo que as de menor porte e com cefalofolio mais arredondado formam um grupo polifilético. Rhizoprionodon acutus aparece como monofilético à família Sphyrnidae quando incluído fora do grupo externo. As árvores de consenso (Strictus, Semi-strictus, Majority-rule e Adams) apresentam os mesmos resultados. Os caracteres cranianos, sensoriais e de dentículo dérmico são os que mais suportaram os diversos clados. Após a revisão taxonômica da família, a filogenia com base em morfologia apresentou-se mais consistente e clara, embora controversa aos dados moleculares.