Biology subjects

Do, C.

Publications and source records attributed to Do, C..

22 records · Page 2Linked to original sources

bioRxiv2020

CloudASM: an ultra-efficient cloud-based pipeline for mapping allele-specific DNA methylation

SummaryMethods for quantifying the imbalance in CpG methylation between alleles genome-wide have been described but their algorithmic time complexity is quadratic and their practical use requires painstaking attention to infrastructure choice, implementation, and execution. To solve this problem, we developed CloudASM, a scalable, ultra-efficient, turn-key, portable pipeline on Google Cloud Computing (GCP) that uses a novel pipeline manager and GCPs serverless enterprise data warehouse. Availability and ImplementationCloudASM is freely available in the GitHub repository https://github.com/TyckoLab/CloudASM and a sample dataset and its results are also freely available at https://console.cloud.google.com/storage/browser/cloudasm. Contactemmanuel.dumont@hmh-cdi.org Supplementary informationNone.

bioinformatics↗

bioRxiv2019

Allele-specific DNA methylation is increased in cancers and its dense mapping in normal plus neoplastic cells increases the yield of disease-associated regulatory SNPs

BackgroundMapping of allele-specific DNA methylation (ASM) can be a post-GWAS strategy for localizing regulatory sequence polymorphisms (rSNPs). However, the advantages of this approach, and the mechanisms underlying ASM in normal and neoplastic cells, remain to be clarified. ResultsWe performed whole genome methyl-seq on diverse normal cells and tissues and three types of cancers (multiple myeloma, lymphoma, glioblastoma multiforme). After excluding imprinting, the data pinpointed 15,114 high-confidence ASM differentially methylated regions (DMRs), of which 1,842 contained SNPs in strong linkage disequilibrium or coinciding with GWAS peaks. ASM frequencies were increased 5 to 9-fold in cancers vs. matched normal tissues, due to widespread allele-specific hypomethylation and focal allele-specific hypermethylation in poised chromatin. Cancers showed increased allele switching at ASM loci, but disruptive SNPs in specific classes of CTCF and transcription factor (TF) binding motifs were similarly correlated with ASM in cancer and non-cancer. Rare somatic mutations affecting these same motif classes tracked with de novo ASM in the cancers. Allele-specific TF binding from ChIP-seq was enriched among ASM loci, but most ASM DMRs lacked such annotations, and some were found in otherwise uninformative "chromatin deserts". ConclusionsASM is increased in cancers but occurs by a shared mechanism involving disruptive SNPs in CTCF and TF binding sites in both normal and neoplastic cells. Dense ASM mapping in normal plus cancer samples reveals candidate rSNPs that are difficult to find by other approaches. Together with GWAS data, these rSNPs can nominate specific transcriptional pathways in susceptibility to autoimmune, neuropsychiatric, and neoplastic diseases. Custom genome browser tracks with annotated ASM loci can be viewed at a UCSC browser session hosted by our laboratory (https://bit.ly/tycko-asm)

genomics↗

bioRxiv2018

Genetic variants influence on the placenta regulatory landscape

BackgroundFrom genomic association studies, quantitative trait loci analysis, and epigenomic mapping, it is evident that significant efforts are necessary to define genetic-epigenetic interactions and understand their role in disease susceptibility and progression. For this reason, an analysis of the effects of genetic variation on gene expression and DNA methylation in human placentas at high resolution and whole-genome coverage will have multiple mechanistic and practical implications.\n\nResultsBy producing and analyzing DNA sequence variation (n=303), DNA methylation (n=303) and mRNA expression data (n=80) from placentas from healthy women, we investigate the regulatory landscape of the human placenta and offer analytical approaches to integrate different types of genomic data and address some potential limitations of current platforms. We distinguish two profiles of interaction between expression and DNA methylation, revealing linear or bimodal effects, reflecting differences in genomic context, transcription factor recruitment, and possibly cell subpopulations.\n\nConclusionsThese findings help to clarify the interactions of genetic, epigenetic, and transcriptional regulatory mechanisms in normal human placentas. They also provide strong evidence for genotype-driven modifications of transcription and DNA methylation in normal placentas. In addition to these mechanistic implications, the data and analytical methods presented here will improve the interpretability of genome-wide and epigenome-wide association studies for human traits and diseases that involve placental functions.\n\nAuthor summaryThe placenta is a critical organ playing multiple roles including oxygen and metabolite transfer from mother to fetus, hormone production, and vascular perfusion. With this study, we aimed to deliver a placenta-specific regulatory map based on a combination of publicly available and newly generated data. To complete this reference, we obtained genotype information (n=303), DNA methylation (n=303) and expression data (n=80) for placentas from healthy women. Our analysis of methylation and expression quantitative trait loci (QTLs) and correlations between methylation and expression data were designed to identify fundamental associations between genome, transcriptome, and epigenome in this key fetal organ. The results provide high-resolution genetic and epigenetic maps specific to the placenta based on a representative ethnically diverse cohort. As interest and efforts are growing to better understand the etiology of placental disease and the impact of the environment on placental function these data will provide a reference and enhance future investigations.

genomics↗

bioRxiv2018

Deep Learning Based BCI Control of a Robotic Service Assistant Using Intelligent Goal Formulation

As autonomous service robots become more affordable and thus available for the general public, there is a growing need for user-friendly interfaces to control these systems. Control interfaces typically get more complicated with increasing complexity of the robotic tasks and the environment. Traditional control modalities as touch, speech or gesture commands are not necessarily suited for all users. While non-expert users can make the effort to familiarize themselves with a robotic system, paralyzed users may not be capable of controlling such systems even though they need robotic assistance most. In this paper, we present a novel framework, that allows these users to interact with a robotic service assistant in a closed-loop fashion, using only thoughts. The system is composed of several interacting components: non-invasive neuronal signal recording and co-adaptive deep learning which form the brain-computer interface (BCI), high-level task planning based on referring expressions, navigation and manipulation planning as well as environmental perception. We extensively evaluate the BCI in various tasks, determine the performance of the goal formulation user interface and investigate its intuitiveness in a user study. Furthermore, we demonstrate the applicability and robustness of the system in real world scenarios, considering fetch-and-carry tasks and tasks involving human-robot interaction. As our results show, the system is capable of adapting to frequent changes in the environment and reliably accomplishes given tasks within a reasonable amount of time. Combined with high-level planning using referring expressions and autonomous robotic systems, interesting new perspectives open up for non-invasive BCI-based human-robot interactions.

neuroscience↗