Wellcome Trust Centre for Human Genetics

About: Wellcome Trust Centre for Human Genetics is a facility organization based out in Oxford, United Kingdom. It is known for research contribution in the topics: Population & Genome-wide association study. The organization has 2122 authors who have published 4269 publications receiving 433899 citations.

MicroRNA-10b and breast cancer metastasis

Abstract: Arising from: L. Ma, J. Teruya-Feldstein & R. A. Weinberg , 682–688 (2007)10.1038/nature06174 ; Ma et al. reply MicroRNAs regulate messenger RNA expression but are frequently dysregulated in tumours. Ma et al.1 report that overexpression of microRNA-10b (miR-10b) initiates invasion and metastasis in models of breast cancer and that its expression in primary breast carcinomas correlates with clinical progression. We tested this in patients with primary breast cancer, of whom 92 had nodal metastases at diagnosis and 127 were node-negative. We found no significant association between miR-10b levels and metastasis or prognosis. Although we concede that miR-10b may have a biological effect in a few cells at the growing edge of a tumour, we believe that it is unlikely to correlate in whole tumour samples with clinical progression.

Predictor correlation impacts machine learning algorithms

Abstract: Motivation: The advent of high-throughput genomics has produced studies with large numbers of predictors (e.g. genome-wide association, microarray studies). Machine learning algorithms (MLAs) are a computationally efficient way to identify phenotype-associated variables in high-dimensional data. There are important results from mathematical theory and numerous practical results documenting their value. One attractive feature of MLAs is that many operate in a fully multivariate environment, allowing for small-importance variables to be included when they act cooperatively. However, certain properties of MLAs under conditions common in genomic-related data have not been well-studied—in particular, correlations among predictors pose a problem. Results: Using extensive simulation, we showed considering correlation within predictors is crucial in making valid inferences using variable importance measures (VIMs) from three MLAs: random forest (RF), conditional inference forest (CIF) and Monte Carlo logic regression (MCLR). Using a case–control illustration, we showed that the RF VIMs—even permutation-based—were less able to detect association than other algorithms at effect sizes encountered in complex disease studies. This reduction occurred when ‘causal’ predictors were correlated with other predictors, and was sharpest when RF tree building used the Gini index. Indeed, RF Gini VIMs are biased under correlation, dependent on predictor correlation strength/number and over-trained to random fluctuations in data when tree terminal node size was small. Permutation-based VIM distributions were less variable for correlated predictors and are unbiased, thus may be preferred when predictors are correlated. MLAs are a powerful tool for high-dimensional data analysis, but well-considered use of algorithms is necessary to draw valid conclusions. Contact: kristin.nicodemus@well.ox.ac.uk Supplementary information: Supplementary data are available at Bioinformatics online.

Extensive regulation of the non‐coding transcriptome by hypoxia: role of HIF in releasing paused RNApol2

Abstract: Hypoxia is central to both ischaemic and neoplastic diseases. However, the non-coding transcriptional response to hypoxia is largely uncharacterized. We undertook integrated genomic analyses of both non-coding and coding transcripts using massively parallel sequencing and interfaced this data with pan-genomic analyses of hypoxia-inducible factor (HIF) and RNApol2 binding in hypoxic cells. These analyses revealed that all classes of RNA are profoundly regulated by hypoxia and implicated HIF as a major direct regulator of both the non-coding and coding transcriptome, acting predominantly through release of pre-bound promoter-paused RNApol2. These findings indicate that the transcriptional response to hypoxia is substantially more extensive than previously considered.

A single-cell atlas of the human substantia nigra reveals cell-specific pathways associated with neurological disorders

Abstract: We describe a human single-nuclei transcriptomic atlas for the substantia nigra (SN), generated by sequencing approximately 17,000 nuclei from matched cortical and SN samples. We show that the common genetic risk for Parkinson’s disease (PD) is associated with dopaminergic neuron (DaN)-specific gene expression, including mitochondrial functioning, protein folding and ubiquitination pathways. We identify a distinct cell type association between PD risk and oligodendrocyte-specific gene expression. Unlike Alzheimer’s disease (AD), we find no association between PD risk and microglia or astrocytes, suggesting that neuroinflammation plays a less causal role in PD than AD. Beyond PD, we find associations between SN DaNs and GABAergic neuron gene expression and multiple neuropsychiatric disorders. Conditional analysis reveals that distinct neuropsychiatric disorders associate with distinct sets of neuron-specific genes but converge onto shared loci within oligodendrocytes and oligodendrocyte precursors. This atlas guides our aetiological understanding by associating SN cell type expression profiles with specific disease risk. The substantia nigra is important in neurological disease, particularly movement disorders. Here the authors provide a single cell transcriptomic atlas for the human substantia nigra.