Andrew W. Dowsey

Image analysis tools and emerging algorithms for expression proteomics.

Since their origins in academic endeavours in the 1970s, computational analysis tools have matured into a number of established commercial packages that underpin research in expression proteomics. In this paper we describe the image analysis pipeline for the established 2‐DE technique of protein separation, and by first covering signal analysis for MS, we also explain the current image analysis workflow for the emerging high‐throughput ‘shotgun’ proteomics platform of LC coupled to MS (LC/MS). The bioinformatics challenges for both methods are illustrated and compared, whereas existing commercial and academic packages and their workflows are described from both a users and a technical perspective. Attention is given to the importance of sound statistical treatment of the resultant quantifications in the search for differential expression. Despite wide availability of proteomics software, a number of challenges have yet to be overcome regarding algorithm accuracy, objectivity and automation, generally due to deterministic spot‐centric approaches that discard information early in the pipeline, propagating errors. We review recent advances in signal and image analysis algorithms in 2‐DE, MS, LC/MS and Imaging MS. Particular attention is given to wavelet techniques, automated image‐based alignment and differential analysis in 2‐DE, Bayesian peak mixture models, and functional mixed modelling in MS, and group‐wise consensus alignment methods for LC/MS.

Automated image alignment for 2D gel electrophoresis in a high-throughput proteomics pipeline

MOTIVATIONnThe quest for high-throughput proteomics has revealed a number of challenges in recent years. Whilst substantial improvements in automated protein separation with liquid chromatography and mass spectrometry (LC/MS), aka shotgun proteomics, have been achieved, large-scale open initiatives such as the Human Proteome Organization (HUPO) Brain Proteome Project have shown that maximal proteome coverage is only possible when LC/MS is complemented by 2D gel electrophoresis (2-DE) studies. Moreover, both separation methods require automated alignment and differential analysis to relieve the bioinformatics bottleneck and so make high-throughput protein biomarker discovery a reality. The purpose of this article is to describe a fully automatic image alignment framework for the integration of 2-DE into a high-throughput differential expression proteomics pipeline.nnnRESULTSnThe proposed method is based on robust automated image normalization (RAIN) to circumvent the drawbacks of traditional approaches. These use symbolic representation at the very early stages of the analysis, which introduces persistent errors due to inaccuracies in modelling and alignment. In RAIN, a third-order volume-invariant B-spline model is incorporated into a multi-resolution schema to correct for geometric and expression inhomogeneity at multiple scales. The normalized images can then be compared directly in the image domain for quantitative differential analysis. Through evaluation against an existing state-of-the-art method on real and synthetically warped 2D gels, the proposed analysis framework demonstrates substantial improvements in matching accuracy and differential sensitivity. High-throughput analysis is established through an accelerated GPGPU (general purpose computation on graphics cards) implementation.nnnAVAILABILITYnSupplementary material, software and images used in the validation are available at http://www.proteomegrid.org/rain/.

MR image-based geometric and hemodynamic investigation of the right coronary artery with dynamic vessel motion.

The aim of this study was to develop a fully subject-specific model of the right coronary artery (RCA), including dynamic vessel motion, for computational analysis to assess the effects of cardiac-induced motion on hemodynamics and resulting wall shear stress (WSS). Vascular geometries were acquired in the right coronary artery (RCA) of a healthy volunteer using a navigator-gated interleaved spiral sequence at 14 time points during the cardiac cycle. A high temporal resolution velocity waveform was also acquired in the proximal region. Cardiac-induced dynamic vessel motion was calculated by interpolating the geometries with an active contour model and a computational fluid dynamic (CFD) simulation with fully subject-specific information was carried out using this model. The results showed the expected variation of vessel radius and curvature throughout the cardiac cycle, and also revealed that dynamic motion of the right coronary artery consequent to cardiac motion had significant effects on instantaneous WSS and oscillatory shear index. Subject-specific MRI-based CFD is feasible and, if scan duration could be shortened, this method may have potential as a non-invasive tool to investigate the physiological and pathological role of hemodynamics in human coronary arteries.

Guidelines for reporting the use of gel image informatics in proteomics

655 1LGC, Teddington, Middlesex, UK. 2International Graduate School of Arts and Sciences, Yokohama City University, Tsurumi-ku, Yokohama, Kanagawa, Japan. 3Facultad de Farmacia, Universidad San Pablo-CEU, Campus Montepríncipe, Boadilla del Monte, Madrid, Spain. 4Bioproduct Research and Development, Lilly Research Laboratories, Lilly Technology Centre, Indianapolis, Indiana, USA. 5Department of Protein Analytical Chemistry, Genentech Inc., South San Francisco, California, USA. 6Pharmaceutical Sciences Research Division, King’s College London, London, UK. 7School of Biomedical Sciences, University of Ulster, Coleraine, Co. Londonderry, UK. 8Aalen University, Aalen, Germany. 9William Harvey Research Institute, Barts & The London School of Medicine and Dentistry, Queen Mary University of London, Charterhouse Square, London, UK. 10Max-Planck-Institute for Dynamics of Complex Technical Systems, Magdeburg, Germany. 11Lilly UK, Speke, Liverpool, UK. 12European Bioinformatics Institute, Hinxton, UK (chris.taylor@ebi.ac.uk).

Motion-compensated MR valve imaging with COMB tag tracking and super-resolution enhancement

MR imaging of the heart valve leaflets is a challenging problem due to their large movements throughout the cardiac cycle. This paper presents a motion-compensated imaging approach with COMB tagging for valve imaging. It involves an automatic method for tracking the full 3D motion of the valve plane so as to provide a motion-tracked acquisition scheme. Super-resolution enhancement is then applied to the slice-select direction so that the partial volume effect is minimised. In vivo results have shown that in terms of slice positioning, the method has equivalent accuracy to that of a manual approach whilst being quicker and more consistent. The use of multiple parallel COMB tags will permit adaptive imaging that follows tissue motion. This will have significant implications for quantification of myocardial perfusion and tracking anatomy, functions that are traditionally difficult in MRI.

Informatics and Statistics for Analyzing 2-D Gel Electrophoresis Images

Despite recent progress in shotgun peptide separation by integrated liquid chromatography and mass spectrometry (LC/MS), proteome coverage and reproducibility are still limited with this approach and obtaining enough replicate runs for biomarker discovery is a challenge. For these reasons, recent research demonstrates that there is a continuing need for protein separation by two-dimensional gel electrophoresis (2-DE). However, with traditional 2-DE informatics, the digitized images are reduced to symbolic data through spot detection and quantification before proteins are compared for differential expression by spot matching. Recently, a more robust and automated paradigm has emerged where gels are directly aligned in the image domain before spots are detected across the whole image set as a whole. In this chapter, we describe the methodology for both approaches and discuss the pitfalls present when reasoning statistically about the differential protein expression discovered.

Cardiovascular magnetic resonance tagging of the right ventricular free wall for the assessment of long axis myocardial function in congenital heart disease

BackgroundRight ventricular ejection fraction (RV-EF) has traditionally been used to measure and compare RV function serially over time, but may be a relatively insensitive marker of change in RV myocardial contractile function. We developed a cardiovascular magnetic resonance (CMR) tagging-based technique with a view to rapid and reproducible measurement of RV long axis function and applied it in patients with congenital heart disease.MethodsWe studied 84 patients: 56 with repaired Tetralogy of Fallot (rTOF); 28 with atrial septal defect (ASD): 13 with and 15 without pulmonary hypertension (RV pressure > 40 mmHG by echocardiography). For comparison, 20 healthy controls were studied. CMR acquisitions included an anatomically defined four chamber cine followed by a cine gradient echo-planar sequence in the same plane with a labelling pre-pulse giving a tag line across the basal myocardium. RV tag displacement was measured with automated registration and tracking of the tag line together with standard measurement of RV-EF.ResultsMean RV displacement was higher in the control (26 ± 3 mm) than in rTOF (16 ± 4 mm) and ASD with pulmonary hypertension (18 ± 3 mm) groups, but lower than in the ASD group without (30 ± 4 mm), P < 0.001. The technique was reproducible with inter-study bias ± 95% limits of agreement of 0.7 ± 2.7 mm. While RV-EF was lower in rTOF than in controls (49 ± 9% versus 57 ± 6%, P < 0.001), it did not differ between either ASD group and controls.ConclusionsMeasurements of RV long axis displacement by CMR tagging showed more differences between the groups studied than did RV-EF, and was reproducible, quick and easy to apply. Further work is needed to assess its potential use for the detection of longitudinal changes in RV myocardial function.

The Future of Large-Scale Collaborative Proteomics

The postgenomics era has witnessed a rapid change in biological methods for knowledge elucidation and pharmacological approaches to biomarker discovery. Differential expression of proteins in health and disease holds the key to early diagnosis and accelerated drug discovery. This approach, however, has also brought an explosion of data complexity not mirrored by existing progress in proteome informatics. It has become apparent that the task is greater than that can be tackled by individual laboratories alone and large-scale open collaborations of the new human proteome organization (HUPO) have highlighted major challenges concerning the integration and cross-validation of results across different laboratories. This paper describes the state-of-the-art proteomics workflows (two-dimensional gel electrophoresis, liquid chromatography, and mass spectrometry) and their utilization by the participants of the HUPO initiatives towards comprehensive mapping of the brain, liver, and plasma proteomes. Particular emphasis is given to the limitations of the underlying data analysis techniques for large-scale collaborative proteomics. Emerging paradigms including statistical data normalization, direct image registration, spectral libraries, and high-throughput computation with Web-based bioinformatics services are discussed. It is envisaged that these methods will provide the basis for breaking the bottleneck of large-scale automated proteome mapping and biomarker discovery.

Motion-Compensated MR valve imaging with COMB tag tracking and super-resolution enhancement

Understanding the morphology and function of heart valves is important to the study of underlying causes of heart failure. Existing techniques such as those based on echocardiography are limited by the relatively low signal-to-noise ratio (SNR), attenuation artefacts, and restricted access. The alternative of cardiovascular MR imaging offers versatility and accuracy in 3D localisation, but is hampered by large movements of the valves throughout the cardiac cycle. This paper presents a motion-compensated adaptive imaging approach for MR valve imaging. To illustrate its clinical potential, 3D motion of the aortic valve plane is first captured through a single breath-hold COMB tag pre-scan and then tracked in real-time with an automatic method based on multi-resolution image registration. Motion-compensated coverage of the aortic valve is then acquired prospectively, thus allowing its clear 3D reconstruction and visualisation. To provide isotropic voxel coverage of the imaging volume, retrospective projection onto convex sets (POCS) super-resolution enhancement is applied to the slice-select direction. In vivo results demonstrate the effectiveness of the proposed motion-compensation and super-resolution schemes for depicting the structure of the valve leaflets throughout the cardiac cycle. The proposed method fundamentally changes the way MR imaging is performed by transforming it from a spatially to materially localised imaging method. This also has important implications for quantifying blood flow and myocardial perfusion, as well as tracking anatomy and function of the heart.

Cardiac-motion compensated MR imaging and strain analysis of ventricular trabeculae

In conventional CMR, bulk cardiac motion causes target structures to move in and out of the static acquisition plane. Due to the partial volume effect, accurate localisation of subtle features through the cardiac cycle, such as the trabeculae and papillary muscles, is difficult. This problem is exacerbated by the short acquisition window necessary to avoid motion blur and ghosting, especially during early systole. This paper presents an adaptive imaging approach with COMB multi-tag tracking that follows true 3D motion of the myocardium so that the same tissue slice is imaged throughout the cine acquisition. The technique is demonstrated with motion-compensated multi-slice imaging of ventricles, which allows for tracked visualisation and analysis of the trabeculae and papillary muscles for the first time. This enables novel in-vivo measurement of circumferential and radial strain for trabeculation and papillary muscle contractility. These statistics will facilitate the evaluation of diseases such as mitral valve insufficiency and ischemic heart disease. The adaptive imaging technique will also have significant implications for CMR in general, including motion-compensated quantification of myocardial perfusion and blood flow, and motion-correction of sequences with long acquisition windows.