Mark Eckley - Academia.edu (original) (raw)

Papers by Mark Eckley

ascl, Dec 1, 2013

WND-CHARM quantitatively analyzes morphologies of galaxy mergers and associate galaxies by their ... more WND-CHARM quantitatively analyzes morphologies of galaxy mergers and associate galaxies by their morphology. It computes a large set (up to ~2700) of image features for each image based on the WND-CHARM algorithm. It can then split the images into training and test sets and classify them. The software extracts the image content descriptor from raw images, image transforms, and compound image transforms. The most informative features are then selected, and the feature vector of each image is used for classification and similarity measurement using Fisher discriminant scores and a variation of Weighted Nearest Neighbor analysis. WND-CHARM&#39;s results comparable favorably to the performance of task-specific algorithms developed for tested datasets. The simple user interface allows researchers who are not knowledgeable in computer vision methods and have no background in computer programming to apply image analysis to their data.

Bio-image informatics is the systematic application of image analysis algorithms to large image d... more Bio-image informatics is the systematic application of image analysis algorithms to large image datasets to provide an objective method for accurately and consistently scoring image data. Within this field, pattern recognition (PR) is a form of supervised machine learning where the computer identifies relevant patterns in groups (classes) of images after being trained on examples. Rather than segmentation, image-specific algorithms or adjustable parameter sets, PR relies on extracting a common set of image descriptors (features) from the entire image to determine similarities and differences between image classes. Gross morphology can be the only available description of biological systems prior to their molecular characterization, but these descriptions can be subjective and qualitative. In principle, generalized PR can provide an objective and quantitative characterization of gross morphology, thus providing a means of computationally defining morphological biomarkers. In this study, we investigated the potential of a pattern recognition approach to a problem traditionally addressed using genetic or biochemical biomarkers. Often these molecular biomarkers are unavailable for investigating biological processes that are not well characterized, such as the initial steps of stem cell differentiation. Here we use a general contrast technique combined with generalized PR software to detect subtle differences in cellular morphology present in early differentiation events in murine embryonic stem cells (mESC) induced to differentiate by the overexpression of selected transcription factors. Without the use of reporters, or a priori knowledge of the relevant morphological characteristics, we identified the earliest differentiation event (3 days), reproducibly distinguished eight morphological trajectories, and correlated morphological trajectories of 40 mESC clones with previous micro-array data. Interestingly, the six transcription factors that caused the greatest morphological divergence from an ESC-like state were previously shown by expression profiling to have the greatest influence on the expression of downstream genes.

Molecular Biology of the Cell, Jul 1, 2003

The dynein activator dynactin is a multiprotein complex with distinct microtubule-and cargobindin... more The dynein activator dynactin is a multiprotein complex with distinct microtubule-and cargobinding domains. The cargo-binding domain contains a short, actin-like filament of the actinrelated protein Arp1, a second actin-related protein, Arp11, and conventional actin. The length of this filament is invariant in dynactin isolated from multiple species and tissues, suggesting that activities that regulate Arp1 polymerization are important for dynactin assembly. Arp11 is present in a protein complex localized at the pointed end of the Arp1 minifilament, whereas actin capping protein (CapZ) is present at the barbed end. Either might cooperate with conventional actin to cap Arp1. We tested the ability of Arp11 to interact with conventional actin and found it could coassemble. Like Arp1, cytosolic Arp11 is found only in dynactin, suggesting that Arp11 and free cytosolic actin do not interact significantly. Recombinant Arp11 and Arp1 were demonstrated to interact by coprecipitation. We developed an in vivo assay for Arp11-Arp1 interaction based on previous observations that Arp1 forms filamentous assemblies when overexpressed in cultured cells. Arp11 significantly decreases the formation of these organized Arp1 assemblies. Finally, this assay was used to confirm the identity of a putative Arp11 homolog in Drosophila melanogaster.

Source Code for Biology and Medicine, Jul 8, 2008

Background: Biological imaging is an emerging field, covering a wide range of applications in bio... more Background: Biological imaging is an emerging field, covering a wide range of applications in biological and clinical research. However, while machinery for automated experimenting and data acquisition has been developing rapidly in the past years, automated image analysis often introduces a bottleneck in high content screening. Methods: Wndchrm is an open source utility for biological image analysis. The software works by first extracting image content descriptors from the raw image, image transforms, and compound image transforms. Then, the most informative features are selected, and the feature vector of each image is used for classification and similarity measurement. Results: Wndchrm has been tested using several publicly available biological datasets, and provided results which are favorably comparable to the performance of task-specific algorithms developed for these datasets. The simple user interface allows researchers who are not knowledgeable in computer vision methods and have no background in computer programming to apply image analysis to their data. Conclusion: We suggest that wndchrm can be effectively used for a wide range of biological image analysis tasks. Using wndchrm can allow scientists to perform automated biological image analysis while avoiding the costly challenge of implementing computer vision and pattern recognition algorithms.

Journal of Cell Biology, Oct 18, 1999

The multiprotein complex, dynactin, is an integral part of the cytoplasmic dynein motor and is re... more The multiprotein complex, dynactin, is an integral part of the cytoplasmic dynein motor and is required for dynein-based motility in vitro and in vivo. In living cells, perturbation of the dynein-dynactin interaction profoundly blocks mitotic spindle assembly, and inhibition or depletion of dynein or dynactin from meiotic or mitotic cell extracts prevents microtubules from focusing into spindles. In interphase cells, perturbation of the dynein-dynactin complex is correlated with an inhibition of ER-to-Golgi movement and reorganization of the Golgi apparatus and the endosome-lysosome system, but the effects on microtubule organization have not previously been defined. To explore this question, we overexpressed a variety of dynactin subunits in cultured fibroblasts. Subunits implicated in dynein binding have effects on both microtubule organization and centrosome integrity. Microtubules are reorganized into unfocused arrays. The pericentriolar components, ␥ tubulin and dynactin, are lost from centrosomes, but pericentrin localization persists. Microtubule nucleation from centrosomes proceeds relatively normally, but microtubules become disorganized soon thereafter. Overexpression of some, but not all, dynactin subunits also affects endomembrane localization. These data indicate that dynein and dynactin play important roles in microtubule organization at centrosomes in fibroblastic cells and provide new insights into dynactin-cargo interactions.

Biochemistry, Aug 25, 1987

In a previous paper we presented evidence for the existence of at least two a,-acid glycoprotein ... more In a previous paper we presented evidence for the existence of at least two a,-acid glycoprotein (AGP) genes in the mouse. One of the cDNA clones characterized in those studies was used to isolate several unique AGP genomic clones. In these studies we present the complete sequence of one of the mouse AGP genes. The sequence analysis includes 595 base pairs (bp) 5' to the site of initiation of transcription and 135 bp 3' to the polyadenylation signal. This mouse AGP gene, designated AGP-1, has six exons, a structure similar to those of the A G P genes in rats and humans. Analysis of the sequence has revealed a number of potential regulatory sites. These include a run of alternating purine-pyrimidine bases [(GT),] at +2890 to +2945, flanked by three potential glucocorticoid receptor binding sites within intron 5. Two of these TGTTCT a t +3069 to +3074 and +3082 to +3087 flank the (GT), track at its 3' end, and one, which is oriented in the opposite direction (AGAACA), at +2771 to +2776 flanks the track a t its 5' end. A longer version of the glucocorticoid receptor site, GGGTACAATGTGTCCT, has been located in the 5' flanking region of the gene (-94 to-79); the sequence AGAACA is another potential glucocorticoid receptor site oriented in the opposite direction and located at-127 to-122. This entire region, from-146 to-42, in the mouse has a strong homology (-85%) to the 5' flanking region of the rat AGP gene, which contains a 78-bp fragment (-1 20 to-42) that represents the minimal sequence required for glucocorticoid regulation. A sequence of 38 nucleotides (-22 to +16) that is homologous to similarly located sequences previously observed in three human acute-phase proteins has also been identified. We suggest that this sequence may represent an acute-phase protein regulatory element.

Chromosoma, May 1, 2001

The inner centromere protein (INCENP), which has previously been described in chicken, frog and m... more The inner centromere protein (INCENP), which has previously been described in chicken, frog and mouse, is required for correct chromosome segregation and cytokinesis. We have identified the human INCENP gene by library screening and reverse transcription-polymerase chain reaction (RT-PCR) and localized it to chromosomal region 11q12. HsINCENP is a singlecopy gene that consists of 17 exons and covers 25 kb of genomic DNA. The gene is expressed at highest levels in the colon, testis and prostate, consistent with its likely role in cell proliferation. HsINCENP encodes a highly basic protein of 915 amino acids that localizes to metaphase chromosomes and to the mitotic spindle and equatorial cortex at anaphase. Recently we showed that INCENP is stockpiled in a complex with the Aurora-B/XAIRK2 kinase in Xenopus eggs. Here we demonstrate that, consistent with such an interaction, the two proteins colocalize on human metaphase chromosomes. Levels of Aurora-B are increased in several human cancers, and we show here that HsINCENP protein levels are also significantly increased in several colorectal cancer cell lines. Abbreviations aa: amino acid(s) • ACA: anticentromere antibodies • AIRK: Aurora/Ipl1-related kinase • bp: base pair(s) • cAMP-PKA: cyclic AMP-dependent protein kinase A • DAPI: 4′,6-diamidino-2-phenylindole • EST: expressed sequence tag • INCENP: inner centromere protein • ORF: open reading frame • MAP kinase: mitogen-activated protein kinase • PBS: phosphate-buffered saline • RT-PCR: reverse transcription-polymerase chain reaction

Journal of Cell Biology, Mar 9, 1998

INCENP is a tightly bound chromosomal protein that transfers to the spindle midzone at the metaph... more INCENP is a tightly bound chromosomal protein that transfers to the spindle midzone at the metaphase/anaphase transition. Here, we show that an INCENP truncation mutant (INCENP 382-839) associates with microtubules but does not bind to chromosomes, and coats the entire spindle throughout mitosis. Furthermore, an INCENP truncation mutant (INCENP 43-839) previously shown not to transfer to the spindle at anaphase (

The Journals of Gerontology, Dec 5, 2017

In this study, we describe a morphological biomarker that detects multiple discrete subpopulation... more In this study, we describe a morphological biomarker that detects multiple discrete subpopulations (or "age-states") at several chronological ages in a population of nematodes (Caenorhabditis elegans). We determined the frequencies of three healthy adult states and the timing of the transitions between them across the lifespan. We used short-lived and long-lived strains to confirm the general applicability of the state classifier and to monitor state progression. This exploration revealed healthy and unhealthy states, the former being favored in long-lived strains and the latter showing delayed onset. Short-lived strains rapidly transitioned through the putative healthy state. We previously found that age-matched animals in different age-states have distinct transcriptome profiles. We isolated animals at the beginning and end of each identified state and performed microarray analysis (principal component analysis, relative sample to sample distance measurements, and gene set enrichment analysis). In some comparisons, chronologically identical individuals were farther apart than morphologically identical individuals isolated on different days. The age-state biomarker allowed assessment of aging in a novel manner, complementary to chronological age progression. We found hsp70 and some small heat shock protein genes are expressed later in adulthood, consistent with the proteostasis collapse model.

ACM Transactions on Applied Perception, Feb 1, 2010

We describe a method for automated recognition of painters and schools of art based on their sign... more We describe a method for automated recognition of painters and schools of art based on their signature styles, and studied the computer-based perception of visual art. Paintings of nine artists, representing three different schools of art-impressionism, surrealism and abstract expressionism-were analyzed using a large set of image features and image transforms. The computed image descriptors were assessed using Fisher scores, and the most informative features were used for the classification and similarity measurements of paintings, painters, and schools of art. Experimental results show that the classification accuracy when classifying paintings into nine painter classes is 77%, and the accuracy of associating a given painting with its school of art is 91%. An interesting feature of the proposed method is its ability to automatically associate different artists that share the same school of art in an unsupervised fashion. The source code used for the image classification and image similarity described in this paper is available for free download.

The post genomic era introduced the need to define single gene functions within biological pathwa... more The post genomic era introduced the need to define single gene functions within biological pathways. A systems biology approach can be realized by automating image acquisition and phenotype classification. While machinery for automated data acquisition have been developing rapidly in the past years, the main bottleneck remains the effectiveness of the computer vision algorithms. Here we describe a fully automated process for finding phenotype similarities within a dataset acquired from an RNAi screen. The source code for the algorithms is available for free download.

Springer eBooks, 2008

Oxygen delivery requires that Red Blood Cells (RBCs) must be deformable to pass through the micro... more Oxygen delivery requires that Red Blood Cells (RBCs) must be deformable to pass through the microcirculation. Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by abnormal extracellular deposition of b-amyloid peptide (Ab) and neuronal loss. We have analyzed RBC morphology in blood from subjects with AD and found that >15% of the RBCs are elongated as compared to 5.9% in normal controls (p<0.0001). To determine whether these morphology changes can be associated with the greater exposure of RBCs to Ab in AD subjects, we investigated the in vitro effect of Ab fibrils on blood. Morphological analysis of RBCs treated with Ab 1-40 or Ab 1-42 fibrils show 8.6% or 11.1% elongated cells, respectively. In contrast, only 2.9% or 1.3% of RBCs are elongated when blood is treated with buffer or mock fibrils generated from Ab 42-1. Elongated RBCs are expected to be less deformable. This prediction is consistent with our earlier studies showing impaired deformability of RBCs treated with Ab fibrils. An additional factor previously reported by us, expected to impair the flow of RBCs through the microcirculation is their adherence to endothelial cells (ECs) when Ab 1-40 fibrils are bound to either RBCs or ECs. This factor would be more pronounced in AD subjects with elevated levels of Ab on the vasculature. These results suggest that Ab interactions with RBCs in AD subjects can result in impaired oxygen transport and delivery, which will have important implications for AD.

Pattern Recognition Letters, Aug 1, 2008

We describe a multipurpose image classifier that can be applied to a wide variety of image classi... more We describe a multipurpose image classifier that can be applied to a wide variety of image classification tasks without modifications or fine-tuning, and yet provide classification accuracy comparable to state-of-the-art task-specific image classifiers. The proposed image classifier first extracts a large set of 1025 image features including polynomial decompositions, high contrast features, pixel statistics, and textures. These features are computed on the raw image, transforms of the image, and transforms of transforms of the image. The feature values are then used to classify test images into a set of pre-defined image classes. This classifier was tested on several different problems including biological image classification and face recognition. Although we cannot make a claim of universality, our experimental results show that this classifier performs as well or better than classifiers developed specifically for these image classification tasks. Our classifier's high performance on a variety of classification problems is attributed to (i) a large set of features extracted from images; and (ii) an effective feature selection and weighting algorithm sensitive to specific image classification problems. The algorithms are available for free download from openmicroscopy.org.

Medical & Biological Engineering & Computing, Jul 31, 2008

New technology for automated biological image acquisition has introduced the need for effective b... more New technology for automated biological image acquisition has introduced the need for effective biological image analysis methods. These algorithms are constantly being developed by pattern recognition and machine vision experts, who tailor general computer vision techniques to the specific needs of biological imaging. However, computer scientists do not always have access to biological image datasets that can be used for computer vision research, and biologist collaborators who can assist in defining the biological questions are not always available. Here we propose a publicly available benchmark suite of biological image datasets that can be used by machine vision experts for developing and evaluating biological image analysis methods. The suite represents a set of practical real-life imaging problems in biology, and offers examples of organelles, cells and tissues, imaged at different magnifications and different contrast techniques. All datasets are available for free download at http://ome.grc.nia.nih.gov/iicbu2008. Keywords Biological imaging; image analysis; image datasets; benchmarks Computer vision algorithms are typically developed by pattern recognition and signal processing experts. In order to develop, test and evaluate the performance of novel algorithms, computer scientists and engineers must have access to datasets of biological images that represent actual biological questions. However, in many real-life cases machine vision specialists do not have convenient access to biological images, and do not work closely with biologists, who are better able to define the biological problems and direct the research towards

PLOS Computational Biology, Nov 24, 2010

The increasing prevalence of automated image acquisition systems is enabling new types of microsc... more The increasing prevalence of automated image acquisition systems is enabling new types of microscopy experiments that generate large image datasets. However, there is a perceived lack of robust image analysis systems required to process these diverse datasets. Most automated image analysis systems are tailored for specific types of microscopy, contrast methods, probes, and even cell types. This imposes significant constraints on experimental design, limiting their application to the narrow set of imaging methods for which they were designed. One of the approaches to address these limitations is pattern recognition, which was originally developed for remote sensing, and is increasingly being applied to the biology domain. This approach relies on training a computer to recognize patterns in images rather than developing algorithms or tuning parameters for specific image processing tasks. The generality of this approach promises to enable data mining in extensive image repositories, and provide objective and quantitative imaging assays for routine use. Here, we provide a brief overview of the technologies behind pattern recognition and its use in computer vision for biological and biomedical imaging. We list available software tools that can be used by biologists and suggest practical experimental considerations to make the best use of pattern recognition techniques for imaging assays.

Molecular Biology of the Cell, May 30, 2001

Journal of Machine Vision and Applications, Jun 29, 2011

In this work we explored class separability in feature spaces built on extended representations o... more In this work we explored class separability in feature spaces built on extended representations of pixel planes (EPP) produced using scale pyramid, subband pyramid, and image transforms. The image transforms included Chebyshev, Fourier, wavelets, gradient and Laplacian; we also utilized transform combinations, including Fourier, Chebyshev and wavelets of the gradient transform, as well as Fourier of the Laplacian transform. We demonstrate that all three types of EPP promote class separation. We also explored the effect of EPP on suboptimal feature libraries, using only textural features in one case and only Haralick features in another. The effect of EPP was especially clear for these suboptimal libraries, where the transform-based representations were found to increase separability to a greater extent than scale or subband pyramids. EPP can be particularly useful in new applications where optimal features have not yet been developed.

Cytometry Part A, Mar 29, 2012

We present results from machine classification of melanoma biopsies sectioned and stained with he... more We present results from machine classification of melanoma biopsies sectioned and stained with hematoxylin/eosin (H&E) on tissue micro-arrays (TMA). The four stages of melanoma progression were represented by seven tissue types, including benign nevus, primary tumors with radial and vertical growth patterns (stage I) and four secondary metastatic tumors: subcutaneous (stage II), lymph node (stage III), gastrointestinal and soft tissue (stage IV). Our experiment setup comprised 14,208 image samples based on 164 TMA cores. In our experiments we constructed an HE color space by digitally deconvolving the RGB images into separate H (hematoxylin) and E (eosin) channels. We also compared three different classifiers: Weighted Neighbor Distance (WND), Radial Basis Functions (RBF), and k-Nearest Neighbors (kNN). We found that the HE color space consistently outperformed other color spaces with all three classifiers, while the different classifiers did not have as large of an effect on accuracy. This showed that a more physiologically relevant representation of color can have a larger effect on correct image interpretation than downstream processing steps. We were able to correctly classify individual fields of view with an average of 96% accuracy when randomly splitting the dataset into training and test fields. We also obtained a classification accuracy of 100% when testing entire cores that were not previously used in training (four random trials with one test core for each of 7 classes, 28 tests total). Because each core corresponded to a different patient, this test more closely mimics a clinically relevant setting where new patients are evaluated based on training with previous cases. The analysis method used in this study contains no parameters or adjustments that are specific to melanoma morphology, suggesting it can be used for analyzing other tissues and phenotypes, as well as potentially different image modalities and contrast techniques.

Journal of Cell Biology, Oct 15, 1993

It has recently been proposed that mitotic chromosomes transport certain cytoskeletal proteins to... more It has recently been proposed that mitotic chromosomes transport certain cytoskeletal proteins to the metaphase plate so that these proteins are able to subsequently participate in the assembly of the anaphase spindle and the cleavage furrow. To understand how such proteins accomplish their dual chromosomahcytoskeletal role, we have begun a molecular and functional analysis of the inner centromere proteins (INCENPs), founder members of the class of "chromosome passenger proteins." eDNA clones encoding the open reading frames of the two chicken IN-CENPs were recovered. The predicted proteins, class I INCENP (96,357 D) and class 1I INCENP (100,931 D) are novel, and differ from each other by the inclusion of a 38-codon insert within the class II coding region. Transient expression of the chicken INCENPs in mammalian cells confirms that the signals and struc-1. Abbreviations used in this paper: AMV, avian myoblastosis virus; IN-CENE inner centromere protein; RT, reverse transcription.

Molecular Biology of the Cell, Oct 1, 2012

Dynactin is an essential part of the cytoplasmic dynein motor that enhances motor processivity an... more Dynactin is an essential part of the cytoplasmic dynein motor that enhances motor processivity and serves as an adaptor that allows dynein to bind cargoes. Much is known about dynactin's interaction with dynein and microtubules, but how it associates with its diverse complement of subcellular binding partners remains mysterious. It has been suggested that cargo specification involves a group of subunits referred to as the "pointed-end complex." We used chemical cross-linking, RNA interference, and protein overexpression to characterize interactions within the pointed-end complex and explore how it contributes to dynactin's interactions with endomembranes. The Arp11 subunit, which caps one end of dynactin's Arp1 filament, and p62, which binds Arp11 and Arp1, are necessary for dynactin stability. These subunits also allow dynactin to bind the nuclear envelope prior to mitosis. p27 and p25, by contrast, are peripheral components that can be removed without any obvious impact on dynactin integrity. Dynactin lacking these subunits shows reduced membrane binding. Depletion of p27 and p25 results in impaired early and recycling endosome movement, but late endosome movement is unaffected, and mitotic spindles appear normal. We conclude that the pointed-end complex is a bipartite structural domain that stabilizes dynactin and supports its binding to different subcellular structures.