Generalization in vision and motor control

Learning is more than memory. It is not simply the building of a look-up table of labelled images, or a phone-directory-like list of motor acts and the corresponding sequences of muscle activation. Central to learning and intelligence is the ability to predict, that is, to generalize to new situations, beyond the memory of specific examples. The key to generalization, in turn, is the architecture of the system, more than the rules of synaptic plasticity. We propose a specific architecture for generalization for both the motor and the visual systems, and argue for a canonical microcircuit underlying visual and motor learning.     

Isolation and characterization of a protochordate histocompatibility locus

Histocompatibility--the ability of an organism to distinguish its own cells and tissue from those of another--is a universal phenomenon in the Metazoa. In vertebrates, histocompatibility is a function of the immune system controlled by a highly polymorphic major histocompatibility complex (MHC), which encodes proteins that target foreign molecules for immune cell recognition. The association of the MHC and immune function suggests an evolutionary relationship between metazoan histocompatibility and the origins of vertebrate immunity. However, the MHC of vertebrates is the only functionally characterized histocompatibility system; the mechanisms underlying this process in non-vertebrates are unknown. A primitive chordate, the ascidian Botryllus schlosseri, also undergoes a histocompatibility reaction controlled by a highly polymorphic locus. Here we describe the isolation of a candidate gene encoding an immunoglobulin superfamily member that, by itself, predicts the outcome of histocompatibility reactions. This is the first non-vertebrate histocompatibility gene described, and may provide insights into the evolution of vertebrate adaptive immunity.     

Diversity and life-history traits of wild bees (Insecta: Hymenoptera) in intensive agricultural landscapes in the Rolling Pampa,Argentina

The decline of bees is a major concern due to their vital role in pollinating many crops and wild plants. Some regions in South America, and especially the Pampas, are amongst those parts of the world where stressors of bee populations have been little studied. The Pampas has been intensively transformed for agriculture, being presently one of the most productive areas of agricultural commodities in the world. Here, we aim to provide first insights on the taxonomic and functional composition of bee assemblages in the Rolling Pampa, the most intensively managed part of the Pampas. Soybean (herbicide-tolerant genetically modified varieties) is the predominant crop in this region. Bees were sampled with coloured pan traps. Sampling points were located on field margins in either the cropped or the semi-natural grassland area of a farmland site devoted to annual cropping. A total of 2384 individuals were caught, representing 33 taxa [mainly (morpho)species]. The subgenus Lasioglossum (Dialictus) largely dominated captures (78% of the total abundance) and was relatively abundant in the entire study area, suggesting that some species are likely to reach their ecological requirements in cropped areas. No-till fields and field margins may provide large areas for these below-ground nesting species, while their polylectic food preferences allow them to collect pollen on a wide range of plant species, among them possibly soybean. On the contrary, the richness and the abundance of other taxa were higher in the semi-natural area than in the cropped area. Among them, above-ground nesting or oil-collecting species, which have more specialised nesting and floral requirements, were highly associated with the semi-natural area. Our findings highlight the large dominance of L. (Dialictus) species in this highly intensively managed landscape, and the urgent need of preserving semi-natural habitats to maintain species-rich and functionally diverse bee communities in the Pampas.     

Computing texture boundaries from images

Recent computational and psychological theories of human texture vision assert that texture discrimination is based on first-order differences in geometric and luminance attributes of texture elements, called 'textons'. Significant differences in the density, orientation, size, or contrast of line segments or other small features in an image have been shown to cause immediate perception of texture boundaries. However, the psychological theories, which are based on the perception of synthetic images composed of lines and symbols, neglect two important issues. First, how can textons be computed from grey-level images of natural scenes? And second, how, exactly, can texture boundaries be found? Our analysis of these two issues has led to an algorithm that is fully implemented and which successfully detects boundaries in natural images. We propose that blobs computed by a centre-surround operator are useful as texture elements, and that a simple non-parametric statistic can be used to compare local distributions of blob attributes to locate texture boundaries. Although designed for natural images, our computation agrees with some psychophysical findings, in particular, those of Adelson and Bergen (described in the preceding article), which cast doubt on the hypothesis that line segment crossings or termination points are textons.     

A parallel algorithm for real-time computation of optical flow

The precise management of two-dimensional field of velocities from time-varying two-dimensional images is impossible in general. It is, however, possible to compute suitable 'optical flows' that are qualitatively similar to the velocity field in most cases. We describe a simple, parallel algorithm that computes an optical flow from sequences of real images, which is consistent with human psychophysics and suggests plausible physiological models. In particular, our algorithm runs on a Connection Machine supercomputer in close-to-real time. It shows several of the same 'illusions' that are perceived by humans. A natural physiological implementation of the model is consistent with data from cortical areas V1 and MT.     

Prediction of central nervous system embryonal tumour outcome based on gene expression.

Embryonal tumours of the central nervous system (CNS) represent a heterogeneous group of tumours about which little is known biologically, and whose diagnosis, on the basis of morphologic appearance alone, is controversial. Medulloblastomas, for example, are the most common malignant brain tumour of childhood, but their pathogenesis is unknown, their relationship to other embryonal CNS tumours is debated, and patients' response to therapy is difficult to predict. We approached these problems by developing a classification system based on DNA microarray gene expression data derived from 99 patient samples. Here we demonstrate that medulloblastomas are molecularly distinct from other brain tumours including primitive neuroectodermal tumours (PNETs), atypical teratoid/rhabdoid tumours (AT/RTs) and malignant gliomas. Previously unrecognized evidence supporting the derivation of medulloblastomas from cerebellar granule cells through activation of the Sonic Hedgehog (SHH) pathway was also revealed. We show further that the clinical outcome of children with medulloblastomas is highly predictable on the basis of the gene expression profiles of their tumours at diagnosis.     

Tumour biology: herceptin acts as an anti-angiogenic cocktail

Malignant tumours secrete factors that enable them to commandeer their own blood supply (angiogenesis), and blocking the action of these factors can inhibit tumour growth. But because tumours may become resistant to treatments that target individual angiogenic factors by switching over to other angiogenic molecules, a cocktail of multiple anti-angiogenic agents should be more effective. Here we show that herceptin, a monoclonal antibody against the cell-surface receptor HER2 (for human epidermal growth factor receptor-2; ref. 4), induces normalization and regression of the vasculature in an experimental human breast tumour that overexpresses HER2 in mice, and that it works by modulating the effects of different pro- and anti-angiogenic factors. As a single agent that acts against multiple targets, herceptin, or drugs like it, may offer a simple alternative to combination anti-angiogenic treatments.     

Pathology: cancer cells compress intratumour vessels

The delivery of therapeutic drugs to solid tumours may be impaired by structural and functional abnormalities in blood and lymphatic vessels. Here we provide evidence that proliferating cancer cells cause intratumour vessels to compress and collapse. By reducing this compressive mechanical force and opening vessels, cytotoxic cancer treatments have the potential to increase blood perfusion, thereby improving drug delivery.