Reference Alignment Research Articles

A word-oriented approach to alignment validation

Multiple sequence alignment at the level of whole proteomes requires a high degree of automation, precluding the use of traditional validation methods such as manual curation. Since evolutionary models are too general to describe the history of each residue in a protein family, there is no single algorithm/model combination that can yield a biologically or evolutionarily optimal alignment. We propose a 'shotgun' strategy where many different algorithms are used to align the same family, and the best of these alignments is then chosen with a reliable objective function. We present WOOF, a novel 'word-oriented' objective function that relies on the identification and scoring of conserved amino acid patterns (words) between pairs of sequences. Tests on a subset of reference protein alignments from BAliBASE showed that WOOF tended to rank the (manually curated) reference alignment highest among 1060 alternative (automatically generated) alignments for a majority of protein families. Among the automated alignments, there was a strong positive relationship between the WOOF score and similarity to the reference alignment. The speed of WOOF and its independence from explicit considerations of three-dimensional structure make it an excellent tool for analyzing large numbers of protein families. On request from the authors.

2P105 単粒子解析における高速unbiased alignment法(蛋白質 E) 計測・解析の方法論))

2P105 単粒子解析における高速unbiased alignment法(蛋白質 E) 計測・解析の方法論))

The influence of gapped positions in multiple sequence alignments on secondary structure prediction methods

All currently leading protein secondary structure prediction methods use a multiple protein sequence alignment to predict the secondary structure of the top sequence. In most of these methods, prior to prediction, alignment positions showing a gap in the top sequence are deleted, consequently leading to shrinking of the alignment and loss of position-specific information. In this paper we investigate the effect of this removal of information on secondary structure prediction accuracy. To this end, we have designed SymSSP, an algorithm that post-processes the predicted secondary structure of all sequences in a multiple sequence alignment by (i) making use of the alignment's evolutionary information and (ii) re-introducing most of the information that would otherwise be lost. The post-processed information is then given to a new dynamic programming routine that produces an optimally segmented consensus secondary structure for each of the multiple alignment sequences. We have tested our method on the state-of-the-art secondary structure prediction methods PHD, PROFsec, SSPro2 and JNET using the HOMSTRAD database of reference alignments. Our consensus-deriving dynamic programming strategy is consistently better at improving the segmentation quality of the predictions compared to the commonly used majority voting technique. In addition, we have applied several weighting schemes from the literature to our novel consensus-deriving dynamic programming routine. Finally, we have investigated the level of noise introduced by prediction errors into the consensus and show that predictions of edges of helices and strands are half the time wrong for all the four tested prediction methods.

Daily megavoltage CT image guidance in the treatment of localized prostate cancer

Daily megavoltage CT image guidance in the treatment of localized prostate cancer

Inverse parametric sequence alignment

We consider the inverse parametric sequence alignment problem, where a sequence alignment, called a reference alignment, is given and the task is to determine parameter values such that the reference alignment is optimal for those parameter values. The goal is to produce biologically meaningful alignments, by using reference alignments as a training set. We describe a O( mnlog n)-time algorithm for inverse global alignment without gap penalty and a O( mnlog m) time algorithm for global alignment with gap penalty, where m, n ( n⩽ m) are the lengths of input strings. We then discuss algorithms for local alignment and multiple sequence alignment.

Stereotactic IMRT for prostate cancer: Setup accuracy of a new stereotactic body localization system

The purpose of this work is to prospectively assess the setup accuracy that can be achieved with a stereotactic body localizer (SBL) in immobilizing patients for stereotactic intensity-modulated radiotherapy (IMRT) for prostate cancer. By quantifying this important factor and target mobility in the SBL, we expect to provide a guideline for selecting planning target volume margins for stereotactic treatment planning. We analyzed data from 40 computed tomography (CT) studies (with slice thickness of 3 mm) involving 10 patients with prostate cancer. Each patient had four sets of CT scans during the course of radiotherapy. For the purpose of this study, all four sets of CT scans were obtained with the patients immobilized in a customized body pillow formed by vacuum suction. Unlike other immobilization devices, this system consists not only of a customized body pillow, but also of a fixation sheet used to suppress patient respiratory motion, a stereotactic body frame to provide stereotaxy, and a carbon fiber base board to which both the body cushion and the frame are affixed. We identified four bony landmarks and measured their coordinates in the stereotactic body frame on each set of CT scans. The displacements of the bony landmarks from their corresponding positions on the simulation scan (first CT scan) were analyzed in three dimensions in terms of overall, systematic, and random categories. The initial planned isocenter was also marked on the patients' skin with fiducials for each CT study. The distance from each bony landmark to the fiducial-based isocenter was measured and compared among the four sets of CT scans. The deviations in distances were also compared to those measured from the landmarks to the stereotactic frame center, in order to determine the effectiveness of the rigid body frame in positioning patients with prostate cancer. Target inter-fraction motion in this system was also studied for five patients by measuring the deviations in distances from the target geometric center to the bony landmarks. Our results showed that the overall setup accuracy had standard deviations (SDs) of 2.58 mm, 2.41 mm, and 3.51 mm in lateral (LAT), anterior-posterior (AP), and superior-inferior (SI) directions, respectively. The random component had SDs of 1.72 mm, 2.06 mm, and 2.79 mm, and the systematic component showed SDs of 0.92 mm, –0.27 mm, and 0.90 mm in these three directions. In terms of three-dimensional vector, the mean displacement over 116 measurements was 3.0 mm with an SD of 1.29 mm. Compared to the rigid reference, the skin-mark-based reference was less reliable for patient repositioning in terms of reproducing known bony landmark positions. The mean target mobility relative to the bony landmarks was , and in the AP, LAT, and SI directions, respectively. In conclusion, the body immobilization system has the ability to immobilize prostate cancer patients with satisfactory setup accuracy for fractionated extracranial stereotactic radiotherapy. A rigid frame system serves as a reliable alignment reference in terms of repositioning patients into the planning position, while skin-based reference showed larger deviations in repositioning patients. PACS number: 87.53Ly

Stereotactic IMRT for prostate cancer: Setup accuracy of a new stereotactic body localization system

The purpose of this work is to prospectively assess the setup accuracy that can be achieved with a stereotactic body localizer (SBL) in immobilizing patients for stereotactic intensity‐modulated radiotherapy (IMRT) for prostate cancer. By quantifying this important factor and target mobility in the SBL, we expect to provide a guideline for selecting planning target volume margins for stereotactic treatment planning. We analyzed data from 40 computed tomography (CT) studies (with slice thickness of 3 mm) involving 10 patients with prostate cancer. Each patient had four sets of CT scans during the course of radiotherapy. For the purpose of this study, all four sets of CT scans were obtained with the patients immobilized in a customized body pillow formed by vacuum suction. Unlike other immobilization devices, this system consists not only of a customized body pillow, but also of a fixation sheet used to suppress patient respiratory motion, a stereotactic body frame to provide stereotaxy, and a carbon fiber base board to which both the body cushion and the frame are affixed. We identified four bony landmarks and measured their coordinates in the stereotactic body frame on each set of CT scans. The displacements of the bony landmarks from their corresponding positions on the simulation scan (first CT scan) were analyzed in three dimensions in terms of overall, systematic, and random categories. The initial planned isocenter was also marked on the patients' skin with fiducials for each CT study. The distance from each bony landmark to the fiducial‐based isocenter was measured and compared among the four sets of CT scans. The deviations in distances were also compared to those measured from the landmarks to the stereotactic frame center, in order to determine the effectiveness of the rigid body frame in positioning patients with prostate cancer. Target inter‐fraction motion in this system was also studied for five patients by measuring the deviations in distances from the target geometric center to the bony landmarks. Our results showed that the overall setup accuracy had standard deviations (SDs) of 2.58 mm, 2.41 mm, and 3.51 mm in lateral (LAT), anterior‐posterior (AP), and superior‐inferior (SI) directions, respectively. The random component had SDs of 1.72 mm, 2.06 mm, and 2.79 mm, and the systematic component showed SDs of 0.92 mm, –0.27 mm, and 0.90 mm in these three directions. In terms of three‐dimensional vector, the mean displacement over 116 measurements was 3.0 mm with an SD of 1.29 mm. Compared to the rigid reference, the skin‐mark‐based reference was less reliable for patient repositioning in terms of reproducing known bony landmark positions. The mean target mobility relative to the bony landmarks was 2.22±3.45 mm,0.17±1.11 mm, and 0.11±2.69 mm in the AP, LAT, and SI directions, respectively. In conclusion, the body immobilization system has the ability to immobilize prostate cancer patients with satisfactory setup accuracy for fractionated extracranial stereotactic radiotherapy. A rigid frame system serves as a reliable alignment reference in terms of repositioning patients into the planning position, while skin‐based reference showed larger deviations in repositioning patients.PACS number: 87.53Ly

Testing homology with Contact Accepted mutatiOn (CAO): a contact-based Markov model of protein evolution

Point Accepted Mutation (PAM) is the Markov model of amino acid replacements in proteins introduced by Dayhoff and her co-workers (Dayhoff et al., 1978). The PAM matrices and other matrices based on the PAM model have been widely accepted as the standard scoring system of protein sequence similarity in protein sequence alignment tools. Here, we present Contact Accepted mutatiOn (CAO), a Markov model of protein residue contact mutations. The CAO model simulates the interchanging of structurally defined side-chain contacts, and introduces additional structural information into protein sequence alignments. Therefore, similarities between structurally conserved sequences can be detected even without apparent sequence similarity. CAO has been benchmarked on the homstrad database and a subset of the cath database, by comparing sequence alignments with reference alignments derived from structural superposition. CAO yields scores that reflect coherently the structural quality of sequence alignments, which has implications particularly for homology modelling and threading techniques.

A Systematic Comparison of Various Statistical Alignment Models

We present and compare various methods for computing word alignments using statistical or heuristic models. We consider the five alignment models presented in Brown, Della Pietra, Della Pietra, and Mercer (1993), the hidden Markov alignment model, smoothing techniques, and refinements. These statistical models are compared with two heuristic models based on the Dice coefficient. We present different methods for combining word alignments to perform a symmetrization of directed statistical alignment models. As evaluation criterion, we use the quality of the resulting Viterbi alignment compared to a manually produced reference alignment. We evaluate the models on the German-English Verbmobil task and the French-English Hansards task. We perform a detailed analysis of various design decisions of our statistical alignment system and evaluate these on training corpora of various sizes. An important result is that refined alignment models with a first-order dependence and a fertility model yield significantly better results than simple heuristic models. In the Appendix, we present an efficient training algorithm for the alignment models presented.

Precise [100] crystal orientation determination on ⟨110⟩-oriented silicon wafers

In this paper we propose a novel pre-etch method to determine the [100] direction on the surface of ⟨110⟩ silicon wafers with a diameter of 100 mm for precise bulk etching. A series of circular windows is arranged in an arc with a radius of 48.9 mm, and bulk etched to form hexagonal shapes for the indication of crystal orientation. The hexagons, which have two angles of 109° and four other angles of 125.5°, are surrounded by four {111} vertical planes and two {111} planes inclined 35.5° to the wafer surface. The corners of the hexagons are used as an alignment reference to indicate the [100] direction on the ⟨110⟩ silicon wafers. Using a calculation from the relationship between the circular windows with different diameters of 153, 74 and 35 μm and the circle center distances of 192, 96 and 48 μm, the alignment accuracy can be determined as ±0.11°, ±0.06° and ±0.03° to the projected [100] orientation, respectively. Experimental results also demonstrate the feasibility of accurately aligning long etching slots along the ⟨111⟩ direction. The misalignment has been determined to be 0.02° from 20 experimental samples, much less than the estimated value of ±0.03° on a 100 mm ⟨110⟩ wafer. This simple, accurate and fast alignment technique is applicable to long slot fabrication on ⟨110⟩ wafers with tight geometry tolerance.

An expectation maximization algorithm for training hidden substitution models

We derive an expectation maximization algorithm for maximum-likelihood training of substitution rate matrices from multiple sequence alignments. The algorithm can be used to train hidden substitution models, where the structural context of a residue is treated as a hidden variable that can evolve over time. We used the algorithm to train hidden substitution matrices on protein alignments in the Pfam database. Measuring the accuracy of multiple alignment algorithms with reference to BAliBASE (a database of structural reference alignments) our substitution matrices consistently outperform the PAM series, with the improvement steadily increasing as up to four hidden site classes are added. We discuss several applications of this algorithm in bioinformatics.

BAliBASE (Benchmark Alignment dataBASE): enhancements for repeats, transmembrane sequences and circular permutations.

BAliBASE is specifically designed to serve as an evaluation resource to address all the problems encountered when aligning complete sequences. The database contains high quality, manually constructed multiple sequence alignments together with detailed annotations. The alignments are all based on three-dimensional structural superpositions, with the exception of the transmembrane sequences. The first release provided sets of reference alignments dealing with the problems of high variability, unequal repartition and large N/C-terminal extensions and internal insertions. Here we describe version 2.0 of the database, which incorporates three new reference sets of alignments containing structural repeats, trans-membrane sequences and circular permutations to evaluate the accuracy of detection/prediction and alignment of these complex sequences. BAliBASE can be viewed at the web site http://www-igbmc.u-strasbg. fr/BioInfo/BAliBASE2/index.html or can be downloaded from ftp://ftp-igbmc.u-strasbg.fr/pub/BAliBASE2 /.

A Two-Exposure Technique for Ice-Embedded Samples Successfully Reconstructs the Chlorocruorin Pigment of Sabella spallanzanii at 2.1 nm Resolution

A technique for reconstructing ice-embedded macromolecules from electron micrographs taken at two specimen tilts (±23°) has been used to determine the structure of chlorocruorin isolated from the Polychaete annelid Sabella spallanzanii. Images of individual molecules were extracted in couples from two micrographs of the same field of view so each couple consists of two projections of the same molecule. One couple was used as a fixed reference for alignment. Different references yielded reconstructions with different orientations. These were merged to give a model against which the orientation of 1624 first-exposure images was refined to give a final reconstruction at 2.1 nm resolution. The structure of this hematic pigment, essentially the same as that for Lumbricus terrestris, is a bilayer structure with overall symmetry D6, containing six hollow groups per layer. A hollow group is formed by six globular masses and has approximate threefold symmetry. Other structural elements connect the two layers and the hollow groups in a layer. This non-globin material occupies about 15% of the total molecular volume. The results show that the double-exposure strategy, previously described by some of the authors and tested in computer simulations, performs well in real experiments and could be used to obtain preliminary reconstructions in a semiautomatic way.

Impact of ultrasound and computed tomography prostate volume registration on evaluation of permanent prostate implants

Ultrasound (US)-guided permanent prostate implants typically use US prostate volumes to plan the implant procedure and CT prostate volumes for 3D dosimetric evaluation of the implant. Such a protocol requires that CT and US prostate volumes be registered. We have studied the impact of prostate volume registration on postimplant dosimetry for patients with low-grade prostate cancer treated with combined US and fluoroscopic-guided permanent implants. A US image set was obtained with the patient in the lithotomy position to delineate the prostate volume that was subsequently used for treatment planning. Each plan was customized and optimized to ensure complete coverage of the US prostate volume. After implant, a CT scan was obtained for postimplant dosimetry with the patient lying supine. Sources were localized on CT by interactively creating orthogonal images of small cubes, whose dimensions were slightly larger than the source, to assure unique identification of each seed. Ultrasound and CT 3D surfaces were registered using either (a) the rectal surface and base of the prostate, or (b) the Foley balloon and urethra as the alignment reference. A dose distribution was assigned to the US prostate volume based on the CT source distribution, and the dose-volume histogram (DVH) was calculated. Prostate volumes drawn from US images differ from those drawn from CT images with the CT volumes being typically larger than the US volumes. Urethral registration of the prostate volume based on aligning the prostatic urethra generates a dose distribution that best follows the preimplant plan and is geometrically the preferable choice for dosimetry. The dose distribution and the DVH for the US prostate is sensitive to the mode of registration limiting the ability to determine if acceptable dose coverage has been achieved.

Biological crystals formation studied by HTREM and image analysis

High resolution electron microscopy (HREM) allows the determination of the molecular crystal structure by comparing HREM images with simulated images. Direct comparison was not possible for small crystalline areas such as nanometer-sized particles or for thin crystals with weak image contrast. To overcome these problems, we used numerical image analysis to gain access to the structure informations within these minute crystals. We used this approach for the characterization of the initial mineralization steps during human amelogenesis, chicken bone and human dentine crystals growth.Our method consists in HREM associated with both numerical image analysis and image simulation techniques (EMS) Image analysis was performed using the IMAGIC statistical. Nanoparticle subimages, 128 × 128 pixels in size, were extracted from the original micrographs by interactive selection. A circular mask with a radius of 50 pixels was applied. The mean intensity of each subimage was set to zero and its internal variance was normalized to 10. Double direct alignment procedures were used to align the images in rotation and translation against an alignment reference extracted from the data set. An average image was finally calculated to improve the signal to noise ratio.

SAGA: sequence alignment by genetic algorithm

We describe a new approach to multiple sequence alignment using genetic algorithms and an associated software package called SAGA. The method involves evolving a population of alignments in a quasi evolutionary manner and gradually improving the fitness of the population as measured by an objective function which measures multiple alignment quality. SAGA uses an automatic scheduling scheme to control the usage of 22 different operators for combining alignments or mutating them between generations. When used to optimise the well known sums of pairs objective function, SAGA performs better than some of the widely used alternative packages. This is seen with respect to the ability to achieve an optimal solution and with regard to the accuracy of alignment by comparison with reference alignments based on sequences of known tertiary structure. The general attraction of the approach is the ability to optimise any objective function that one can invent.

A novel laser alignment system for tracking detectors using transparent silicon strip sensors

Modern large-area precision tracking detectors require increasing accuracy of the geometrical alignment over large distances. A novel optical multi-point alignment system has been developed for the muon spectrometer of the ATLAS detector at the Large Hadron Collider. The system uses collimated laser beams as alignment references which are monitored by semi-transparent optical position sensors. The custom designed sensors provide very precise and uniform position information on the order of 1 μm over a wide measurement range. At suitable laser wavelengths, produced by laser diodes, transmission rates above 90% have been achieved which allow one to align more than 30 sensors along one laser beam. With this capability and equipped with integrated readout electronics, the alignment system offers high flexibility for precision applications in a wide range of detector systems.

Cellular basis of EEG slow rhythms: a study of dynamic corticothalamic relationships

A slow oscillation (< 1 Hz) has recently been described in intracellular recordings from the neocortex and thalamus (Steriade et al., 1993c-e). The aim of the present study was to determine the phase relations between cortical and thalamic neuronal activities during the slow EEG oscillation. Intracellular recordings were performed in anesthetized cats from neurons in motor and somatosensory cortical areas, the rostrolateral sector of the reticular (RE) thalamic nucleus, and thalamocortical (TC) cells from ventrolateral (VL) nucleus. The EEG was used as time reference for alignment of activities in different, simultaneously recorded neurons, including dual impalements of cortical cells as well as cortical and TC cells. The spontaneous EEG oscillation was characterized by slowly recurring (0.3-0.9 Hz) sequences of surface-positive (depth-negative) sharp deflections, often followed by oscillatory activity within the frequency range of sleep spindles (7-14 Hz) or at faster frequencies. Cortical and RE cells were similarly hyperpolarized during the depth-positive EEG waves and were depolarized during the depth-negative EEG deflections. In many instances, the cell depolarization was associated with oscillations at the spindle frequency or with tonic firing at rates related to the level of depolarization. TC neurons were hyperpolarized during the depth-positive EEG waves and displayed a series of IPSPs, at the spindle frequencies, during the depth-negative EEG waves. Depending on the membrane potential (Vm), TC cells could fire spike bursts at the onset of the EEG depth-negativity, or their firing could be delayed by subsequent IPSPs. The sequence of spontaneous EEG and cellular events described above also characterized the responses to cortical and thalamic stimulation. Simultaneous intracellular recordings of pairs of cortical cells or cortical and TC cells showed that spontaneous transitions from less synchronized to more synchronized EEG states were marked by a simultaneous hyperpolarization, coincident with an overt depth-positive EEG wave. We conclude that during low-frequency oscillatory states, characteristic of slow-wave sleep, neocortical and thalamic neurons display phase relations that are restricted to narrow time windows, and that synchronization results from a generalized inhibitory phenomenon. Moreover, EEG synchronization is reflected as active inhibition in TC neurons. That this pattern is also present in states of hypersynchronization, such as seizure activity, is shown in the following paper (Steriade and Contreras, 1994).

Alignment of an off-axis parabolic mirror with two parallel He-Ne laser beams

Two parallel laser beams are used to align an off-axis parabolic mirror without alignment telescope and reference flat. In this method the optical axis of an off-axis parabolic mirror is made parallel to the incident laser beams, in the plane of incidence, by measuring directions of the reflected beams and by changing height and orientation of the mirror. Then, the focal point ofthe off-axis parabolic mirror is automatically found where the two reflected beams cross each other. The alignment of the optical axis to the incident beams is done without knowing focal length and off-axis distance of the mirror. Alignment sensitivity is derived both numerically and analytically. When focal length is 457 mm, off-axis distance is 127 mm, and diameter is 178 mm, the off-axis parabolic mirror is aligned to the incident beams with an angular error of less than 3 mrad.

Magnetic alignment and the Poisson alignment reference system

Three distinct metrological operations are necessary to align a free-electron laser (FEL): the magnetic axis must be located, a straight line reference (SLR) must be generated, and the magnetic axis must be related to the SLR. This article begins with a review of the motivation for developing an alignment system that will assure better than 100-μm accuracy in the alignment of the magnetic axis throughout an FEL. The 100-μm accuracy is an error circle about an ideal axis for 300 m or more. The article describes techniques for identifying the magnetic axes of solenoids, quadrupoles, and wiggler poles. Propagation of a laser beam is described to the extent of revealing sources of nonlinearity in the beam. Development of a straight-line reference based on the Poisson line, a diffraction effect, is described in detail. Spheres in a large-diameter laser beam create Poisson lines and thus provide a necessary mechanism for gauging between the magnetic axis and the SLR. Procedures for installing FEL components and calibrating alignment fiducials to the magnetic axes of the components are also described. The Poisson alignment reference system should be accurate to 25 μm over 300 m, which is believed to be a factor-of-4 improvement over earlier techniques. An error budget shows that only 25% of the total budgeted tolerance is used for the alignment reference system, so the remaining tolerances should fall within the allowable range for FEL alignment.