SpliceViNCI: Visualizing the splicing of non-canonical introns through recurrent neural networks.

Most of the current computational models for splice junction prediction are based on the identification of canonical splice junctions. However, it is observed that the junctions lacking the consensus dimers GT and AG also undergo splicing. Identification of such splice junctions, called the non-canonical splice junctions, is also essential for a comprehensive understanding of the splicing phenomenon. This work focuses on the identification of non-canonical splice junctions through the application of a bidirectional long short-term memory (BLSTM) network. Furthermore, we apply a back-propagation-based (integrated gradient) and a perturbation-based (occlusion) visualization techniques to extract the non-canonical splicing features learned by the model. The features obtained are validated with the existing knowledge from the literature. Integrated gradient extracts features that comprise contiguous nucleotides, whereas occlusion extracts features that are individual nucleotides distributed across the sequence.

Using the Chou's 5-steps rule to predict splice junctions with interpretable bidirectional long short-term memory networks.

Neural models have been able to obtain state-of-the-art performances on several genome sequence-based prediction tasks. Such models take only nucleotide sequences as input and learn relevant features on their own. However, extracting the interpretable motifs from the model remains a challenge. This work explores various existing visualization techniques in their ability to infer relevant sequence information learnt by a recurrent neural network (RNN) on the task of splice junction identification. The visualization techniques have been modulated to suit the genome sequences as input. The visualizations inspect genomic regions at the level of a single nucleotide as well as a span of consecutive nucleotides. This inspection is performed based on the modification of input sequences (perturbation based) or the embedding space (back-propagation based). We infer features pertaining to both canonical and non-canonical splicing from a single neural model. Results indicate that the visualization techniques produce comparable performances for branchpoint detection. However, in the case of canonical donor and acceptor junction motifs, perturbation based visualizations perform better than back-propagation based visualizations, and vice-versa for non-canonical motifs. The source code of our stand-alone SpliceVisuL tool is available at https://github.com/aaiitggrp/SpliceVisuL.

To evaluate the fracture resistance of proclined endodontically treated teeth with different post and core systems: In vitro study.

AIM: This study was aimed to evaluate the fracture resistance of proclined endodontically treated teeth with different post and core systems. SETTINGS AND DESIGN: Experimental in vitro study. METHODOLOGY: Eighty extracted maxillary central incisors were selected and decoronated keeping 2 mm of crown ferrule and were endodontically treated. Postspace was prepared retaining 5 mm apical gp using peeso 3. Samples were divided into two groups. In Group 1, Wax pattern fabricated to accommodate different core angulations to be casted with Ni-Cr alloy. In Group 2, ever stick posts were angulated and cemented followed by porcelain fused to metal crown cementation for both the groups. Samples were thermocycled and subjected to the universal testing machine. STATISTICAL ANALYSIS USED: One-way ANOVA and Tukey's post hoc test were used to compare the mean fracture resistance between different angulations in cast post and ever stick posts. Student's paired t-test was used to compare the mean fracture resistance between cast post and everstick posts for each angulations. P value was set at P < 0.05. RESULTS: The fracture resistance was the highest at 20° and lowest at 30° core angulations in both the groups. CONCLUSIONS: Changing core angulation up to 20° can be carried out safely using any of the post systems tested in the study. Core angulations >20°, should be used with caution, especially in patients with abnormal parafunctional habits and occlusal trauma.

In Vivo Mutagenicity Testing of Arylboronic Acids and Esters.

Arylboronic acids and esters (referred to collectively as arylboronic compounds) are commonly used intermediates in the synthesis of pharmaceuticals but pose a challenge for chemical syntheses because they are often positive for bacterial mutagenicity in vitro. As such, arylboronic compounds are then typically controlled to levels that are acceptable for mutagenic impurities, that is, the threshold of toxicological concern (TTC). This study used ICH M7 guidance to design and conduct a testing strategy to investigate the in vivo relevance of the in vitro positive findings of arylboronic compounds. Eight arylboronic compounds representing a variety of chemical scaffolds were tested in Sprague Dawley and/or Wistar rats in the in vivo Pig-a (peripheral blood reticulocytes and mature red blood cells) and/or comet assays (duodenum and/or liver). Five of the eight compounds were also tested in the micronucleus (peripheral blood) assay. The arylboronic compounds tested orally demonstrated high systemic exposure; thus the blood and bone marrow were adequately exposed to test article. One compound was administered intravenously due to formulation stability issues. This investigation showed that arylboronic compounds that were mutagenic in vitro were not found to be mutagenic in the corresponding in vivo assays. Therefore, arylboronic compounds similar to the scaffolds tested in this article may be considered non-mutagenic and managed in accordance with the ICH Q3A/Q3B guidelines. Environ. Mol. Mutagen. 2019. © 2019 Wiley Periodicals, Inc.

SpliceVec: Distributed feature representations for splice junction prediction.

Identification of intron boundaries, called splice junctions, is an important part of delineating gene structure and functions. This also provides valuable insights into the role of alternative splicing in increasing functional diversity of genes. Identification of splice junctions through RNA-seq is by mapping short reads to the reference genome which is prone to errors due to random sequence matches. This encourages identification of splicing junctions through computational methods based on machine learning. Existing models are dependent on feature extraction and selection for capturing splicing signals lying in the vicinity of splice junctions. But such manually extracted features are not exhaustive. We introduce distributed feature representation, SpliceVec, to avoid explicit and biased feature extraction generally adopted for such tasks. SpliceVec is based on two widely used distributed representation models in natural language processing. Learned feature representation in form of SpliceVec is fed to multilayer perceptron for splice junction classification task. An intrinsic evaluation of SpliceVec indicates that it is able to group true and false sites distinctly. Our study on optimal context to be considered for feature extraction indicates inclusion of entire intronic sequence to be better than flanking upstream and downstream region around splice junctions. Further, SpliceVec is invariant to canonical and non-canonical splice junction detection. The proposed model is consistent in its performance even with reduced dataset and class-imbalanced dataset. SpliceVec is computationally efficient and can be trained with user-defined data as well.

Validation of automatic scanning of microscope slides in recovering rare cellular events: application for detection of fetal cells in maternal blood.

OBJECTIVE: Detection of rare fetal cells (FCs) in the maternal circulation could be used for non-invasive prenatal diagnosis. Considering that FCs in maternal blood are present in extremely low frequency, manual scanning is cumbersome, time-consuming, and unsuitable for clinical applications. As an alternative, we optimized a custom-made classifier for automatic detection of FCs. METHODS: Using MetaSystems' automated platform, we developed a robust detection algorithm and validated its efficiency on retrieval of rare XY cells in a pure population of XX cells. Slides were scanned for presence of predefined XY cells after fluorescence in situ hybridization (FISH) and primed in situ labeling (PRINS). Retrieval of FCs was also performed on samples from maternal blood. RESULTS: The efficiency of detection of rare XY cells was 88% using FISH (117/133) in comparison with 78% (53/68) with PRINS. FC frequencies per 1 mL of maternal blood ranged from 3 to 6 FCs in normal pregnancies versus 13 to 21 FCs in Down syndrome pregnancies. CONCLUSION: Automatic scanning was more efficient and consistent than manual scanning for detection of rare FCs and required considerably less operator time. Automatic scanning using FISH is more sensitive than that using PRINS. The study validates automatic scanning retrieval of FCs from maternal blood.

Automated analysis of protein expression and gene amplification within the same cells of paraffin-embedded tumour tissue.

BACKGROUND: The simultaneous detection of protein expression and gene copy number changes in patient samples, like paraffin-embedded tissue sections, is challenging since the procedures of immunohistochemistry (IHC) and Fluorescence in situ Hybridization (FISH) negatively influence each other which often results in suboptimal staining. Therefore, we developed a novel automated algorithm based on relocation which allows subsequent detection of protein content and gene copy number changes within the same cell. METHODS: Paraffin-embedded tissue sections of colorectal cancers were stained for CD133 expression. IHC images were acquired and image coordinates recorded. Slides were subsequently hybridized with fluorescently labeled DNA probes. FISH images were taken at the previously recorded positions allowing for direct comparison of protein expression and gene copy number signals within the same cells/tissue areas. Relocation, acquisition of the IHC and FISH images, and enumeration of FISH signals in the immunophenotyped tumour areas were done in an automated fashion. RESULTS: Automated FISH analysis was performed on 13 different colon cancer samples that had been stained for CD133; each sample was scored for MYC, ZNF217 and Chromosome 6 in CD133 positive and negative glands. From the 13 cases four (31%) showed amplification for the MYC oncogene and seven of 13 (54%) cases were amplified for ZNF217. There was no significant difference between CD133 positive tumour and CD133 negative tumour cells. CONCLUSION: The technique and algorithm presented here enables an easy and reproducible combination of IHC and FISH based on a novel automated algorithm using relocation and automated spot counting.

Automated analysis of protein expression and gene amplification within the same cells of paraffin-embedded tumour tissue.

BACKGROUND: The simultaneous detection of protein expression and gene copy number changes in patient samples, like paraffin-embedded tissue sections, is challenging since the procedures of immunohistochemistry (IHC) and Fluorescence in situ Hybridization (FISH) negatively influence each other which often results in suboptimal staining.Therefore, we developed a novel automated algorithm based on relocation which allows subsequent detection of protein content and gene copy number changes within the same cell. METHODS: Paraffin-embedded tissue sections of colorectal cancers were stained for CD133 expression. IHC images were acquired and image coordinates recorded. Slides were subsequently hybridized with fluorescently labeled DNA probes. FISH images were taken at the previously recorded positions allowing for direct comparison of protein expression and gene copy number signals within the same cells/tissue areas. Relocation, acquisition of the IHC and FISH images, and enumeration of FISH signals in the immunophenotyped tumour areas were done in an automated fashion. RESULTS: Automated FISH analysis was performed on 13 different colon cancer samples that had been stained for CD133; each sample was scored for MYC, ZNF217 and Chromosome 6 in CD133 positive and negative glands. From the 13 cases four (31%) showed amplification for the MYC oncogene and seven of 13 (54%) cases were amplified for ZNF217. There was no significant difference between CD133 positive tumour and CD133 negative tumour cells. CONCLUSION: The technique and algorithm presented here enables an easy and reproducible combination of IHC and FISH based on a novel automated algorithm using relocation and automated spot counting.