Biology subjects

Lee, T.

Publications and source records attributed to Lee, T..

7 recordsLinked to original sources

bioRxiv2018

Regulation of histone methylation by automethylation of PRC2

Polycomb Repressive Complex 2 (PRC2) is a histone methyltransferase whose function is critical for regulating transcriptional repression in many eukaryotes including humans. Its catalytic moiety EZH2 is responsible for the tri-methylation of H3K27 and also undergoes automethylation. Using mass spectroscopic analysis of recombinant human PRC2, we identified three methylated lysine residues (K510, K514, K515) on a disordered but highly conserved loop of EZH2. These lysines were mostly mono- and di-methylated. Either mutation of these lysines or their methylation increases PRC2 histone methyltransferase activity. In addition, mutation of these three lysines in HEK293T cells using CRISPR genome-editing increases global H3K27 methylation levels. EZH2 automethylation occurs intramolecularly (in cis) by methylation of a pseudosubstrate sequence on the flexible loop. This post-translational modification and cis-regulation of PRC2 are analogous to the activation of many protein kinases by autophosphorylation. We therefore propose that EZH2 automethylation provides a way for PRC2 to modulate its histone methyltransferase activity by sensing histone H3 tails, SAM concentration, and perhaps other effectors.

biochemistry↗

bioRxiv2018

Timely double-strand break repair and pathway choice in pericentromeric heterochromatin depend on the histone demethylase dKDM4A

Introduction Introduction Results Discussion Materials Methods References One of the most harmful DNA lesions is a double-strand break (DSB), whose improper repair can lead to formation of aberrant chromosomes linked to cancer and developmental diseases 1. At DSBs, the severed strands of the DNA helix are repaired by a variety of mechanisms. The two major DSB repair pathways are Non-Homologous End Joining (NHEJ) and Homologous Recombination (HR). NHEJ repairs DNA by ligating both ends of the DSB together, often resulting in small insertions and deletions at the break site. HR repair involves more extensive processing of the DSB site, in which 5 to 3 end-resection of the DSB ends results in a single-stranded DNA sequence that invades and perfectly copies homologous sequences to repair the DS ...

cell biology↗

bioRxiv2017

Natural Language Processing for Classification of Acute, Communicable Findings on Unstructured Head CT Reports: Comparison of Neural Network and Non-Neural Machine Learning Techniques

Background and PurposeTo evaluate the accuracy of non-neural and neural network models to classify five categories (classes) of acute and communicable findings on unstructured head computed tomography (CT) reports.\n\nMaterials and MethodsThree radiologists annotated 1,400 head CT reports for language indicating the presence or absence of acute communicable findings (hemorrhage, stroke, hydrocephalus, and mass effect). This set was used to train, develop, and evaluate a non-neural classifier, support vector machine (SVM), in comparisons to two neural network models using convolutional neural networks (CNN) and neural attention model (NAM) Inter-rater agreement was computed using kappa statistics. Accuracy, receiver operated curves, and area under the curve were calculated and tabulated. P-values < 0.05 was significant and 95% confidence intervals were computed.\n\nResultsRadiologist agreement was 86-94% and Cohens kappa was 0.667-0.762 (substantial agreement). Accuracies of the CNN and NAM (range 0.90-0.94) were higher than SVM (range 0.88-0.92). NAM showed relatively equal accuracy with CNN for three classes, severity, mass effect, and hydrocephalus, higher accuracy for the acute bleed class, and lower accuracy for the acute stroke class. AUCs of all methods for all classes were above 0.92.\n\nConclusionsO_LINeural network models (CNN & NAM) generally had higher accuracies compared to the non-neural models (SVM) and have a range of accuracies that comparable to the inter-annotator agreement of three neuroradiologists.\nC_LIO_LIThe NAM method adds ability to hold the algorithm accountable for its classification via heat map generation, thereby adding an auditing feature to this neural network.\nC_LI\n\nAbbreviations

neuroscience↗

bioRxiv2017

Regulating prospero mRNA Stability Determines When Neural Stem Cells Stop Dividing

During Drosophila and vertebrate brain development, the conserved transcription factor Prospero/Prox1 is an important regulator of the transition between proliferation and differentiation. Prospero level is low in neural stem cells and their immediate progeny, but is upregulated in larval neurons and it is unknown how this process is controlled. Here, we use single molecule fluorescent in situ hybridisation to show that larval neurons selectively transcribe a long prospero mRNA isoform containing a 15 kb 3 untranslated region, which is bound in the brain by the conserved RNA-binding protein Syncrip/hnRNPQ. Syncrip binding increases the mRNA stability of the long prospero isoform, which allows an upregulation of Prospero protein production. Our findings highlight a regulatory strategy involving alternative polyadenylation followed by differential post-transcriptional regulation.

developmental biology↗

bioRxiv2017

Imp/Syp Temporal Gradients Govern Decommissioning Of Drosophila Neural Stem Cells

Timing of Drosophila neuroblast decommissioning is controlled in a lineage-specific manner. Following a prepupal ecdysone pulse, the ecdysone receptor and mediator complex cause neuroblasts to shrink. Shrinking is followed by nuclear accumulation of Prospero and cell cycle exit. Only mushroom body (MB) neuroblasts escape early pupal termination. Here, we demonstrate that the opposing temporal gradients of Imp and Syp RNA-binding proteins that govern temporal fate also regulate neuroblast decommissioning. The Imp gradient declines slower in MB neuroblasts so they still express Imp when it is absent from others. The presence of Imp in MB neuroblasts prevents decommissioning partly through inhibiting the mediator complex. Moreover, a timely induction of Imp can protect many non-MB neuroblasts from aging. We also show that the increasing Syp gradient permits Prospero accumulation and neuroblast termination. Together our results reveal that progeny temporal fate and progenitor decommissioning are co-regulated in protracted neuronal lineages.

developmental biology↗

bioRxiv2017

Stem cell intrinsic, Seven-up-triggered temporal factor gradients diversify intermediate neural progenitors

Drosophila type II neuroblasts produce numerous neurons and glia due to transiently amplifying, intermediate neural progenitors (INP). Consecutively born INPs produce morphologically distinct progeny, presumably due to temporal patterning in type II neuroblasts. We therefore profiled type II neuroblasts transcriptome across time. Our results reveal opposing temporal gradients of Imp and Syp RNA-binding proteins (descending and ascending, respectively). Maintaining Imp expression throughout brain development expands the number of neurons/glia with early temporal fate at the expense of cells with late fate. Conversely, precocious upregulation of Syp reduces the number of cells with early fate. Further, we reveal that the transcription factor, Seven-up initiates progression of the Imp/Syp gradients. Interestingly, genetic manipulations that fix Imp or Syp levels still yield progeny with a small range of early fates. We propose that the Seven-up-initiated Imp/Syp gradients create coarse temporal windows within type II neuroblasts to pattern INPs, which subsequently undergo fine-tuned subtemporal patterning.

neuroscience↗

bioRxiv2017

WheatNet: A genome-scale functional network for hexaploid bread wheat, Triticum aestivum

Gene networks provide a system-level overview of genetic organizations and enable the dissection of functional modules underlying complex traits. Here we report the generation of WheatNet, the first genome-scale functional network for T. aestivum and a companion web server (www.inetbio.org/wheatnet). WheatNet was constructed by integrating 20 distinct genomics datasets, including 156,000 wheat-specific co-expression links mined from 1,929 microarray data. A unique feature of WheatNet is that each network node represents either a single gene or a group of genes. We computationally partitioned gene groups mimicking homeologous genes by clustering 99,386 wheat genes, resulting in 20,248 gene groups comprising 63,401 genes and 35,985 individual genes. Thus, WheatNet was constructed using 56,233 nodes, and the final integrated network has 20,230 nodes and 567,000 edges. The edge information of the integrated WheatNet and all 20 component networks are available for download.

genetics↗