Modelling brain representations of abstract concepts

Abstract conceptual representations are critical for human cognition. Despite their importance, key properties of these representations remain poorly understood. Here, we used computational models of distributional semantics to predict multivariate fMRI activity patterns during the activation and contextualization of abstract concepts. We devised a task in which participants had to embed abstract nouns into a story that they developed around a given background context. We found that representations in inferior parietal cortex were predicted by concept similarities emerging in models of distributional semantics. By constructing different model families, we reveal the models’ learning trajectories and delineate how abstract and concrete training materials contribute to the formation of brain-like representations. These results inform theories about the format and emergence of abstract conceptual representations in the human brain.     

HYLAS: program for generating H curves (abstract three-dimensional representations of long DNA sequences)

The computer program HYLAS generates from a standard DNA lettersequence a three-dimensional space curve (H curve) which embodiesthe entire information content of the original nucleotide sequence.The program can display H curves either as two-dimensional (frontand side view) projections or as stereo-pair images. The curvescan be marked at specific nucleotide locations, annotated, rotatedfor observation from any viewing angle, and manipulated forconvenient side-by-side comparisons. Unlike the cumbersome lettersequences, H curves can be drastically condensed in size withoutlosing their ability to reflect the global nucleotide-distributionpattern of the entire DNA sequence. Often, biologically importantloci can be visually identified on the H curves. HYLAS is writtenin FORTRAN with separate mainframe (IBM- VM/CMS) and microcomputer(MS-DOS) versions. It uses the Tektronix-TCS library of graphicsubroutines. Received on October 24, 1988; accepted on July 15, 1989     

Two-dimensional electrophoresis of small-molecular-weight proteins

We have developed a procedure for two-dimensional separation of small-molecular-weight (9000–30,000), acidic (pI 4–6) proteins that allows the use of strips cut from horizontal isoelectric-focusing slab gels for the first dimension, and discontinuous gels containing sodium dodecyl sulfate and high concentrations of urea in the second dimension. This technique facilitates the screening of large numbers of samples and the evaluation of electrofocusing artifacts. We emphasize measures to prevent major problems encountered in the use of this technique, particularly those caused by diffusion and aggregation. We also describe an extension of the method which allows the two-dimensional comparison of many samples in a selected narrow pH zone of interest.     

Two-dimensional gel electrophoresis of proteins for genetic studies in douglas fir (Pseudotsuga menziesii)

A method of two-dimensional gel electrophoresis of proteins from Douglas fir needles is described. Extraction in the presence of sodium dodecyl sulfate (SDS) and 2-mercaptoethanol followed by heating at 100°C produces reliable two-dimensional gels which are convenient for genetic studies. Three genotypes from different geographical origins have been compared: among 225 loci expressed, 22 display regulatory variations and 7 show allelic variations. Thus it is now possible to undertake the genetic study of Douglas fir using this powerful technique.This work was supported in part by Grant ATP PIRDES 508 444 from the INRA-CNRS.     

Two-dimensional infrared spectroscopy displays signatures of structural ordering in peptide aggregates

In the presence of lipid bilayers, the hexapeptide AcWL(5) forms membrane-bound aggregates dominated by beta-secondary structure and is thus a useful model for the onset of peptide aggregation in membrane environments. Two-dimensional infrared (2D IR) spectra in the amide I region for aggregates of AcWL(5) peptides with single isotopic labels provide new insight into the residue-level structural ordering of the aggregated peptides. Separation of spectral information along two axes provides clear indications of the band frequencies and relative intensities, which together are an indication of extended amide coupling networks across structural regions of a particular size. The lowered anharmonicities, relative to free peptide, and the narrow, homogeneous lineshapes of the 2D IR peaks indicate the delocalization of vibrational modes through an ordered structure whose flexibility varies with relative peptide concentration. Crosspeaks between delocalized transitions can be used to estimate the strength of the coupling interactions between neighboring residues. The sensitivity of 2D IR spectra to residue-level structural ordering shows that 2D IR spectroscopy is a powerful technique for probing structures formed during the onset of peptide aggregation.     

Creating abstract topographic representations: Implications for coding,learning and reasoning

Topographic maps are a fundamental and ubiquitous feature of the sensory and motor regions of the brain. There is less evidence for the existence of conventional topographic maps in associational areas of the brain such as the prefrontal cortex and parietal cortex. The existence of topographically arranged anatomical projections is far more widespread and occurs in associational regions of the brain as well as sensory and motor regions: this points to a more widespread existence of topographically organised maps within associational cortex than currently recognised. Indeed, there is increasing evidence that abstract topographic representations may also occur in these regions. For example, a topographic mnemonic map of visual space has been described in the dorsolateral prefrontal cortex and topographically arranged visuospatial attentional signals have been described in parietal association cortex. This article explores how abstract representations might be extracted from sensory topographic representations and subsequently code abstract information. Finally a simple model is presented that shows how abstract topographic representations could be integrated with other information within the brain to solve problems or form abstract associations. The model uses correlative firing to detect associations between different types of stimuli. It is flexible because it can produce correlations between information represented in a topographic or non-topographic coordinate system. It is proposed that a similar process could be used in high-level cognitive operations such as learning and reasoning.     

Sequence signatures of allosteric proteins towards rational design

Allostery is the phenomenon of changes in the structure and activity of proteins that appear as a consequence of ligand binding at sites other than the active site. Studying mechanistic basis of allostery leading to protein design with predetermined functional endpoints is an important unmet need of synthetic biology. Here, we screened the amino acid sequence landscape in search of sequence-signatures of allostery using Recurrence Quantitative Analysis (RQA) method. A characteristic vector, comprised of 10 features extracted from RQA was defined for amino acid sequences. Using Principal Component Analysis, four factors were found to be important determinants of allosteric behavior. Our sequence–based predictor method shows 82.6% accuracy, 85.7% sensitivity and 77.9% specificity with the current dataset. Further, we show that Laminarity-Mean-hydrophobicity representing repeated hydrophobic patches is the most crucial indicator of allostery. To our best knowledge this is the first report that describes sequence determinants of allostery based on hydrophobicity. As an outcome of these findings, we plan to explore possibility of inducing allostery in proteins.     

Two-dimensional polyacrylamide gel electrophoresis of membrane proteins

Two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) is one of the most powerful separation techniques for complex protein solutions. The proteins are first separated according to their isoelectric point, driven by an electric field across a pH gradient. The pH gradient necessary for the separation according to isoelectric point (pL) is usually established by electrophoresing carrier ampholytes prior to and/or concomitantly with the sample. The second dimension is usually a separation according to molecular size. Mostly this separation is performed after complete denaturation of the proteins by sodium dodecyl sulfate and 2-mercaptoethanol (SDS-PAGE). This standard method has considerable disadvantages when relatively hydrophobic membrane proteins are to be separated: cathodic drift, resulting in nonreproducible separation, and the denaturation of the protein, mostly making it impossible to detect native properties of the proteins after separation (e.g., enzymatic activity, antigenicity, intact multimers, and so on). The protocols presented here take care of most of these obstacles. However, there is probably no universal procedure that can guarantee success at first try for any mixture of membrane proteins; some experimentation will be necessary for optimization. Two procedures are each presented: a denaturing (with urea) and a nondenaturing method for IEF in immobilized pH gradient gels using Immobilines, and a denaturing (with SDS and 2-mercaptoethanol) and a nondenaturing technique (with CHAPS) for the second dimension. Essential tips and tricks are presented to keep frustrations of the newcomer at a low level.     

Contextual alignment of biological sequences (Extended abstract)

We present a model of contextual alignment of biological sequences. It is an extension of the classical alignment, in which we assume that the cost of a substitution depends on the surrounding symbols. In this model the cost of transforming one sequence into another depends on the order of editing operations. We present efficient algorithms for calculating this cost, as well as reconstructing (the representation of) all the orders of operations which yield this optimal cost. A precise characterization of the families of linear orders which can emerge this way is given.     

Two-dimensional separation of erythrocyte membrane proteins.

The details of a two-dimensional separation procedure specially designed for the study of erythrocyte membranes are presented. In this highly reproducible method, the membrane proteins are dissolved in sodium dodecyl sulfate and separated first on the basis of charge by isoelectric focusing. The samples are loaded either at the cathode (CIF) or anode (AIF). The CIF samples gave better separation of the acidic proteins, while the AIF was better for the separation of the high molecular weight polypeptides of the erythrocyte. Over 90 discrete polypeptides could be detected with this method in the pH range of 5 to 8. Special attention was given to the higher molecular weight components. For example, six components could be detected within the 90,000 to 100,000 molecular weight range of protein 3, the major membrane protein. A component with the same or very nearly the same molecular weight as spectrin band 2 was detected. It is more basic than spectrin band 2, and both spectrin band 2 and the basic component are readily phosphorylated in the intact cell. However, the phosphorylation of band 2 is cAMP independent while the phosphorylation of the basic component is enhanced by cAMP. In contrast to spectrin, the basic component is not extracted from the membrane with 0.1 mm EDTA, although dilute NaOH will remove it from the membrane. The Ca²⁺-activated transferase of the erythrocyte cytoplasm will not crosslink this component. Calcium does, however, activate the conversion of this component to a lower molecular weight. This high molecular weight basic component has properties attributed to the component labeled 2.1 in Fairbanks' system of nomenclature.     

Two-dimensional electrophoretic analysis of encephalomyocarditis viral proteins

All four capsid proteins of encephalomyocarditis virus and the precursor to two of these were resolved from purified virions with two-dimensional gel electrophoresis. In addition, all of the known stable virus-specific proteins found in infected cells, but not the primary and intermediate precursor proteins, could be resolved with these techniques.     

Two-dimensional reference map of Agrobacterium tumefaciens proteins

Proteomics based on two-dimensional (2-D) gel electrophoresis of proteins followed by spot identification with mass spectrometry is a commonly used method for physiological studies. Physiological proteomics requires 2-D reference maps, on which most of the main proteins are identified. We present a reference map for the bacterial plant pathogen Agrobacterium tumefaciens proteins, which contains more than 300 entries with an isoelectric point (pI) between 4 and 7. The quantitative study of the proteins in the analytical window of the master gel demonstrated unique features, in comparison with other bacteria. In addition, a theoretical analysis of several protein parameters was performed and compared with the experimental results. A comparison of the theoretical molecular weight (MW) of the proteins and their theoretical pI with their vertical and horizontal migration distances, respectively, pointed out the existence of several proteins that strongly diverted from the graph trend-line. These proteins were clearly subjected to post-translational modifications, which changed their pI and/or MW. Additional support for post-translational modifications comes from the identification of multiple spots of the same gene products. Post-translational modifications appear to be more common than expected, at least for soluble proteins, as more than 10% of the proteins were associated with multiple spots.     

Optimized representations and maximal information in proteins

In an effort to quantify loss of information in the processing of protein bioinformatic data, we examine how representations of amino acid sequence and backbone conformation affect the quantity of accessible structural information from local sequence. We propose a method to extract the maximum amount of peptide backbone structural information available in local sequence fragments, given a finite structural data set. Using methods of information theory, we develop an unbiased measure of local structural information that gauges changes in structural distributions when different representations of secondary structure and local sequence are used. We find that the manner in which backbone structure is represented affects the amount and quality of structural information that may be extracted from local sequence. Representations based on virtual bonds capture more structural information from local sequence than a three-state assignment scheme (helix/strand/loop). Furthermore, we find that amino acids show significant kinship with respect to the backbone structural information they carry, so that a collapse of the amino acid alphabet can be accomplished without severely affecting the amount of extractable information. This strategy is critical in optimizing the utility of a limited database of experimentally solved protein structures. Finally, we discuss the similarities within and differences between groups of amino acids in their roles in the local folding code and recognize specific amino acids critical in the formation of local structure.