Artificial intelligence (AI) - the ability of a machine to perform cognitive tasks to achieve a particular goal based on provided data - is revolutionizing and reshaping our health-care systems. The ...current availability of ever-increasing computational power, highly developed pattern recognition algorithms and advanced image processing software working at very high speeds has led to the emergence of computer-based systems that are trained to perform complex tasks in bioinformatics, medical imaging and medical robotics. Accessibility to 'big data' enables the 'cognitive' computer to scan billions of bits of unstructured information, extract the relevant information and recognize complex patterns with increasing confidence. Computer-based decision-support systems based on machine learning (ML) have the potential to revolutionize medicine by performing complex tasks that are currently assigned to specialists to improve diagnostic accuracy, increase efficiency of throughputs, improve clinical workflow, decrease human resource costs and improve treatment choices. These characteristics could be especially helpful in the management of prostate cancer, with growing applications in diagnostic imaging, surgical interventions, skills training and assessment, digital pathology and genomics. Medicine must adapt to this changing world, and urologists, oncologists, radiologists and pathologists, as high-volume users of imaging and pathology, need to understand this burgeoning science and acknowledge that the development of highly accurate AI-based decision-support applications of ML will require collaboration between data scientists, computer researchers and engineers.
Oligonucleotide (oligo)-based FISH has emerged as an important tool for the study of chromosome organization and gene expression and has been empowered by the commercial availability of highly ...complex pools of oligos. However, a dedicated bioinformatic design utility has yet to be created specifically for the purpose of identifying optimal oligo FISH probe sequences on the genome-wide scale. Here, we introduce OligoMiner, a rapid and robust computational pipeline for the genome-scale design of oligo FISH probes that affords the scientist exact control over the parameters of each probe. Our streamlined method uses standard bioinformatic file formats, allowing users to seamlessly integrate new and existing utilities into the pipeline as desired, and introduces a method for evaluating the specificity of each probe molecule that connects simulated hybridization energetics to rapidly generated sequence alignments using supervised machine learning. We demonstrate the scalability of our approach by performing genome-scale probe discovery in numerous model organism genomes and showcase the performance of the resulting probes with diffraction-limited and single-molecule superresolution imaging of chromosomal and RNA targets. We anticipate that this pipeline will make the FISH probe design process much more accessible and will more broadly facilitate the design of pools of hybridization probes for a variety of applications.
•A system for automatic histological grading of prostate cancer was developed.•The classifier was trained based on detailed annotations of six pathologists.•The inter-observer variability was ...addressed by adapting a crowdsourcing approach.•Novel features based on spatial statistics of the nuclei were proposed.•The performance of the classifier was within agreement levels among pathologists.
Display omitted
Prostate cancer (PCa) is a heterogeneous disease that is manifested in a diverse range of histologic patterns and its grading is therefore associated with an inter-observer variability among pathologists, which may lead to an under- or over-treatment of patients. In this work, we develop a computer aided diagnosis system for automatic grading of PCa in digitized histopathology images using supervised learning methods. Our pipeline comprises extraction of multi-scale features that include glandular, cellular, and image-based features. A number of novel features are proposed based on intra- and inter-nuclei properties; these features are shown to be among the most important ones for classification. We train our classifiers on 333 tissue microarray (TMA) cores that were sampled from 231 radical prostatectomy patients and annotated in detail by six pathologists for different Gleason grades. We also demonstrate the TMA-trained classifier’s performance on additional 230 whole-mount slides of 56 patients, independent of the training dataset, by examining the automatic grading on manually marked lesions and randomly sampled 10% of the benign tissue. For the first time, we incorporate a probabilistic approach for supervised learning by multiple experts to account for the inter-observer grading variability. Through cross-validation experiments, the overall grading agreement of the classifier with the pathologists was found to be an unweighted kappa of 0.51, while the overall agreements between each pathologist and the others ranged from 0.45 to 0.62. These results suggest that our classifier’s performance is within the inter-observer grading variability levels across the pathologists in our study, which are also consistent with those reported in the literature.
There is a need for methods that can image chromosomes with genome-wide coverage, as well as greater genomic and optical resolution. We introduce OligoFISSEQ, a suite of three methods that leverage ...fluorescence in situ sequencing (FISSEQ) of barcoded Oligopaint probes to enable the rapid visualization of many targeted genomic regions. Applying OligoFISSEQ to human diploid fibroblast cells, we show how four rounds of sequencing are sufficient to produce 3D maps of 36 genomic targets across six chromosomes in hundreds to thousands of cells, implying a potential to image thousands of targets in only five to eight rounds of sequencing. We also use OligoFISSEQ to trace chromosomes at finer resolution, following the path of the X chromosome through 46 regions, with separate studies showing compatibility of OligoFISSEQ with immunocytochemistry. Finally, we combined OligoFISSEQ with OligoSTORM, laying the foundation for accelerated single-molecule super-resolution imaging of large swaths of, if not entire, human genomes.
Purpose
Most of the existing convolutional neural network (CNN)-based medical image segmentation methods are based on methods that have originally been developed for segmentation of natural images. ...Therefore, they largely ignore the differences between the two domains, such as the smaller degree of variability in the shape and appearance of the target volume and the smaller amounts of training data in medical applications. We propose a CNN-based method for prostate segmentation in MRI that employs statistical shape models to address these issues.
Methods
Our CNN predicts the location of the prostate center and the parameters of the shape model, which determine the position of prostate surface keypoints. To train such a large model for segmentation of 3D images using small data (1) we adopt a stage-wise training strategy by first training the network to predict the prostate center and subsequently adding modules for predicting the parameters of the shape model and prostate rotation, (2) we propose a data augmentation method whereby the training images and their prostate surface keypoints are deformed according to the displacements computed based on the shape model, and (3) we employ various regularization techniques.
Results
Our proposed method achieves a Dice score of 0.88, which is obtained by using both elastic-net and spectral dropout for regularization. Compared with a standard CNN-based method, our method shows significantly better segmentation performance on the prostate base and apex. Our experiments also show that data augmentation using the shape model significantly improves the segmentation results.
Conclusions
Prior knowledge about the shape of the target organ can improve the performance of CNN-based segmentation methods, especially where image features are not sufficient for a precise segmentation. Statistical shape models can also be employed to synthesize additional training data that can ease the training of large CNNs.
Eukaryotic DNA is highly organized within nuclei and this organization is important for genome function. Fluorescent
hybridization (FISH) approaches allow 3D architectures of genomes to be ...visualized. Scalable FISH technologies, which can be applied to whole animals, are needed to help unravel how genomic architecture regulates, or is regulated by, gene expression during development, growth, reproduction, and aging. Here, we describe a multiplexed DNA FISH Oligopaint library that targets the entire
genome at chromosome, three megabase, and 500 kb scales. We describe a hybridization strategy that provides flexibility to DNA FISH experiments by coupling a single primary probe synthesis reaction to dye conjugated detection oligos via bridge oligos, eliminating the time and cost typically associated with labeling probe sets for individual experiments. The approach allows visualization of genome organization at varying scales in all/most cells across all stages of development in an intact animal model system.
Abstract Detection of moving sources over a complicated background is important for several reasons. First is measuring the astrophysical motion of the source. Second is that such motion resulting ...from atmospheric scintillation, color refraction, or astrophysical reasons is a major source of false alarms for image-subtraction methods. We extend the Zackay, Ofek, and Gal-Yam image-subtraction formalism to deal with moving sources. The new method, named the translient (translational transient) detector, applies hypothesis testing between the hypothesis that the source is stationary and that the source is moving. It can be used to detect source motion or to distinguish between stellar variability and motion. For moving source detection, we show the superiority of translient over the proper image subtraction, using the improvement in the receiver-operating characteristic curve. We show that in the small translation limit, translient is an optimal detector of point-source motion in any direction. Furthermore, it is numerically stable, fast to calculate, and presented in a closed form. Efficient transient detection requires both the proper image-subtraction statistics and the translient statistics: When the translient statistic is higher, then the subtraction residual is likely due to motion. We test our algorithm both on simulated data and on real images obtained by the Large Array Survey Telescope. We demonstrate the ability of translient to distinguish between motion and variability, which has the potential to reduce the number of false alarms in transients detection. We provide the translient implementation in Python and MATLAB.
Abstract Asteroid collisions are one of the main processes responsible for the evolution of bodies in the main belt. Using observations of the Dimorphos impact by the DART spacecraft, we estimate how ...asteroid collisions in the main belt may look in the first hours after the impact. If the DART event is representative of asteroid collisions with a ∼1 m sized impactor, then the light curves of these collisions will rise on timescales of about ≳100 s and will remain bright for about 1 hr. Next, the light curve will decay on a few hours' timescale to an intermediate luminosity level in which it will remain for several weeks, before slowly returning to its baseline magnitude. This estimate suffers from several uncertainties due to, e.g., the diversity of asteroid composition, their material strength, and spread in collision velocities. We estimate that the rate of collisions in the main belt with energy similar to or larger than the DART impact is of the order of 7000 yr −1 (±1 dex). The large range is due to the uncertainty in the abundance of ∼1 m sized asteroids. We estimate the magnitude distribution of such events in the main belt, and we show that ∼6% of these events may peak at magnitudes brighter than 21. The detection of these events requires a survey with ≲1 hr cadence and may contribute to our understanding of the asteroids’ size distribution, collisional physics, and dust production. With an adequate survey strategy, new survey telescopes may regularly detect asteroid collisions.
Nuclear compartments are prominent features of 3D chromatin organization, but sequencing depth limitations have impeded investigation at ultra fine-scale. CTCF loops are generally studied at a finer ...scale, but the impact of looping on proximal interactions remains enigmatic. Here, we critically examine nuclear compartments and CTCF loop-proximal interactions using a combination of in situ Hi-C at unparalleled depth, algorithm development, and biophysical modeling. Producing a large Hi-C map with 33 billion contacts in conjunction with an algorithm for performing principal component analysis on sparse, super massive matrices (POSSUMM), we resolve compartments to 500 bp. Our results demonstrate that essentially all active promoters and distal enhancers localize in the A compartment, even when flanking sequences do not. Furthermore, we find that the TSS and TTS of paused genes are often segregated into separate compartments. We then identify diffuse interactions that radiate from CTCF loop anchors, which correlate with strong enhancer-promoter interactions and proximal transcription. We also find that these diffuse interactions depend on CTCF's RNA binding domains. In this work, we demonstrate features of fine-scale chromatin organization consistent with a revised model in which compartments are more precise than commonly thought while CTCF loops are more protracted.
Genome organization involves cis and trans chromosomal interactions, both implicated in gene regulation, development, and disease. Here, we focus on trans interactions in Drosophila, where homologous ...chromosomes are paired in somatic cells from embryogenesis through adulthood. We first address long-standing questions regarding the structure of embryonic homolog pairing and, to this end, develop a haplotype-resolved Hi-C approach to minimize homolog misassignment and thus robustly distinguish trans-homolog from cis contacts. This computational approach, which we call Ohm, reveals pairing to be surprisingly structured genome-wide, with trans-homolog domains, compartments, and interaction peaks, many coinciding with analogous cis features. We also find a significant genome-wide correlation between pairing, transcription during zygotic genome activation, and binding of the pioneer factor Zelda. Our findings reveal a complex, highly structured organization underlying homolog pairing, first discovered a century ago in Drosophila. Finally, we demonstrate the versatility of our haplotype-resolved approach by applying it to mammalian embryos.