Geometric algorithms for the analysis of 2D-electrophoresis gels.

In proteomics, two-dimensional gel electrophoresis (2-DE) is a separation technique for proteins. The resulting protein spots can be identified either by using picking robots and subsequent mass spectrometry or by visual cross inspection of a new gel image with an already analyzed master gel. Difficulties especially arise from inherent noise and irregular geometric distortions in 2-DE images. Aiming at the automated analysis of large series of 2-DE images, or at the even more difficult interlaboratory gel comparisons, the bottleneck is to solve the two most basic algorithmic problems with high quality: Identifying protein spots and computing a matching between two images. For the development of the analysis software CAROl at Freie Universit?t Berlin, we have reconsidered these two problems and obtained new solutions which rely on methods from computational geometry. Their novelties are: 1. Spot detection is also possible for complex regions formed by several "merged" (usually saturated) spots; 2. User-defined landmarks are not necessary for the matching. Furthermore, images for comparison are allowed to represent different parts of the entire protein pattern, which only partially "overlap." The implementation is done in a client server architecture to allow queries via the internet. We also discuss and point at related theoretical questions in computational geometry.     

Computational analysis of alternative splicing using EST tissue information

Expressed sequence tags (ESTs) from normal and tumor tissues have been deposited in public databases. These ESTs and all mRNA sequences were aligned with the human genome sequence using LEADS, Compugen's alternative splicing modeling platform. We developed a novel computational approach to analyze tissue information of aligned ESTs in order to identify cancer-specific alternative splicing and gene segments highly expressed in particular cancers. Several genes, including one encoding a possible pre-mRNA splicing factor, displayed cancer-specific alternative splicing. In addition, multiple candidate gene segments highly expressed in colon cancers were identified.     

Replicating by the clock

The eukaryotic genome is divided into well-defined DNA regions that are programmed to replicate at different times during S phase. Active genes are generally associated with early replication, whereas inactive genes replicate late. This expression pattern might be facilitated by the differential restructuring of chromatin at the time of replication in early or late S phase.     

TNF-alpha associated with extracellular matrix fibronectin provides a stop signal for chemotactically migrating T cells

The migration of T cells into extravascular sites of inflammation is regulated by information derived from the molecular structure of the invaded tissue and from chemokine and cytokine gradients in the context of the extracellular matrix (ECM). Although recent studies have highlighted the role of particular chemoattractants in leukocyte migration, to date little is known about how specific combinations of contextual signals control the migration of leukocytes and their localization at sites of inflammation. Here we studied the interplay between a pleiotropic cytokine, TNF-alpha, and two prototypic chemoattractants, RANTES and stromal cell-derived factor-1alpha (SDF-1alpha), on human CD45RO+ T cells migrating within an ECM-like context. For this purpose, we used a newly constructed three-dimensional gel system designed to follow, in real time, the migration of individual leukocytes along chemotactic gradients in vitro. We found that TNF-alpha, which binds the ECM protein fibronectin and lacks adhesion- and migration-promoting effects of its own, can act as a proadhesive cytokine on T cells exposed to RANTES and SDF-1alpha. Furthermore, fibronectin-complexed TNF-alpha provided anchorage signals to the T cells as they moved directionally along chemoattractive gradients. This effect of TNF-alpha required an intact TNF-alpha receptor II subtype on the migrating T cells. The anchoring effect of TNF-alpha appears to be specific; IL-2, an integrin-activating proadhesive cytokine, does not transmit stoppage signals to T cell migration induced by RANTES. Thus, TNF-alpha present in the ECM at sites of inflammation may function to anchor T cells recruited to these sites by chemotactic signals.     

Computerized analysis of chemotaxis at different stages of bacterial growth

Chemotactic behavior as a function of growth stage in an Escherichia coli strain commonly used for chemotaxis studies was characterized using computerized image analysis. The response and adaptation to saturating, step-like additions of the attractant L-aspartate were measured. Steady-state average tumbling frequency and adaptation time increased nearly twofold during logarithmic phase. In contrast, precision of adaptation, P, defined as the ratio between steady-state tumbling frequencies in the presence and absence of attractant, appeared to be constant throughout growth (P = 1.0 +/- 0.2). The variation of tumble duration over growth was consistent with a hydrodynamic mechanism for tumble termination.     

Predicting and controlling the reactivity of immune cell populations against cancer

Heterogeneous cell populations form an interconnected network that determine their collective output. One example of such a heterogeneous immune population is tumor‐infiltrating lymphocytes (TILs), whose output can be measured in terms of its reactivity against tumors. While the degree of reactivity varies considerably between different TILs, ranging from null to a potent response, the underlying network that governs the reactivity is poorly understood. Here, we asked whether one can predict and even control this reactivity. To address this we measured the subpopulation compositions of 91 TILs surgically removed from 27 metastatic melanoma patients. Despite the large number of subpopulations compositions, we were able to computationally extract a simple set of subpopulation‐based rules that accurately predict the degree of reactivity. This raised the conjecture of whether one could control reactivity of TILs by manipulating their subpopulation composition. Remarkably, by rationally enriching and depleting selected subsets of subpopulations, we were able to restore anti‐tumor reactivity to nonreactive TILs. Altogether, this work describes a general framework for predicting and controlling the output of a cell mixture.     

Multi-node graphs: a framework for multiplexed biological assays.

Multiplex polymerase chain reaction (PCR) is an extension of the standard PCR protocol in which primers for multiple DNA loci are pooled together within a single reaction tube, enabling simultaneous sequence amplification, thus reducing costs and saving time. Potential cost saving and throughput improvements directly depend on the level of multiplexing achieved. Designing reliable and highly multiplexed assays is challenging because primers that are pooled together in a single reaction tube may cross-hybridize, though this can be addressed either by modifying the choice of primers for one or more amplicons, or by altering the way in which DNA loci are partitioned into separate reaction tubes. In this paper, we introduce a new graph formalism called a multi-node graph, and describe its application to the analysis of multiplex PCR scalability. We show, using random multi-node graphs that the scalability of multiplex PCR is constrained by a phase transition, suggesting fundamental limits on efforts to improve the cost-effectiveness and throughput of standard multiplex PCR assays. In particular, we show that when the multiplexing level of the reaction tubes is roughly theta(log (sn)) (where s is the number of primer pair candidates per locus and n is the number of loci to be amplified), then with very high probability we can 'cover' all loci with a valid assignment to one of the tubes in the assay. However, when the multiplexing level of the tube exceeds these bounds, there is no possible cover and moreover the size of the cover drops dramatically. Simulations using a simple greedy algorithm on real DNA data also confirm the presence of this phase transition. Our theoretical results suggest, however, that the resulting phase transition is a fundamental characteristic of the problem, implying intrinsic limits on the development of future assay design algorithms.     

Effects of Neural Morphology and Input Distribution on Synaptic Processing by Global and Focal NMDA-Spikes

Cortical neurons can respond to glutamatergic stimulation with regenerative N-Methyl-D-aspartic acid (NMDA)-spikes. NMDA-spikes were initially thought to depend on clustered synaptic activation. Recent work had shown however a new variety of a global NMDA-spike, which can be generated by randomly distributed inputs. Very little is known about the factors that influence the generation of these global NMDA-spikes, as well the potentially distinct rules of synaptic integration and the computational significance conferred by the two types of NMDA-spikes. Here I show that the input resistance (R_IN) plays a major role in influencing spike initiation; while the classical, focal NMDA-spike depended upon the local (dendritic) R_IN, the threshold of global NMDA-spike generation was set by the somatic R_IN. As cellular morphology can exert a large influence on R_IN, morphologically distinct neuron types can have dissimilar rules for NMDA-spikes generation. For example, cortical neurons in superficial layers were found to be generally prone to global NMDA-spike generation. In contrast, electric properties of cortical layer 5b cells clearly favor focal NMDA-spikes. These differences can translate into diverse synaptic integration rules for the different classes of cortical cells; simulated superficial layers neurons were found to exhibit strong synaptic interactions between different dendritic branches, giving rise to a single integrative compartment mediated by the global NMDA-spike. In these cells, efficiency of postsynaptic activation was relatively little dependent on synaptic distribution. By contrast, layer 5b neurons were capable of true multi-unit computation involving independent integrative compartments formed by clustered synaptic input which could trigger focal NMDA-spikes. In a sharp contrast to superficial layers neurons, randomly distributed synaptic inputs were not very effective in driving firing the layer 5b cells, indicating a possibility for different computation performed by these important cortical neurons.     

Childhood Learning Disabilities and Atypical Dementia: A Retrospective Chart Review

Objective

To further our understanding of the association between self-reported childhood learning disabilities (LDs) and atypical dementia phenotypes (Atypical Dementia), including logopenic primary progressive aphasia (L-PPA), Posterior Cortical Atrophy (PCA), and Dysexecutive-type Alzheimer’s Disease (AD).

Methods

This retrospective case series analysis of 678 comprehensive neuropsychological assessments compared rates of self-reported LD between dementia patients diagnosed with Typical AD and those diagnosed with Atypical Dementia. 105 cases with neuroimaging or CSF data available and at least one neurology follow-up were identified as having been diagnosed by the neuropsychologist with any form of neurodegenerative dementia. These cases were subject to a consensus diagnostic process among three dementia experts using validated clinical criteria for AD and PPA. LD was considered Probable if two or more statements consistent with prior LD were documented within the Social & Developmental History of the initial neuropsychological evaluation.

Results

85 subjects (Typical AD n=68, Atypical AD n=17) were included in the final analysis. In logistic regression models adjusted for age, gender, handedness, education and symptom duration, patients with Probable LD, compared to patients without Probable LD, were significantly more likely to be diagnosed with Atypical Dementia vs. Typical AD (OR 13.1, 95% CI 1.3-128.4). All three of the L-PPA cases reporting a childhood LD endorsed childhood difficulty with language. By contrast, both PCA cases reporting Probable childhood LD endorsed difficulty with attention and/or math.

Conclusions

In people who develop dementia, childhood LD may predispose to atypical phenotypes. Future studies are required to confirm whether atypical neurodevelopment predisposes to regional-specific neuropathology in AD and other dementias.