Andrei L. Turinsky

iRefWeb: interactive analysis of consolidated protein interaction data and their supporting evidence

We present iRefWeb, a web interface to protein interaction data consolidated from 10 public databases: BIND, BioGRID, CORUM, DIP, IntAct, HPRD, MINT, MPact, MPPI and OPHID. iRefWeb enables users to examine aggregated interactions for a protein of interest, and presents various statistical summaries of the data across databases, such as the number of organism-specific interactions, proteins and cited publications. Through links to source databases and supporting evidence, researchers may gauge the reliability of an interaction using simple criteria, such as the detection methods, the scale of the study (high- or low-throughput) or the number of cited publications. Furthermore, iRefWeb compares the information extracted from the same publication by different databases, and offers means to follow-up possible inconsistencies. We provide an overview of the consolidated protein–protein interaction landscape and show how it can be automatically cropped to aid the generation of meaningful organism-specific interactomes. iRefWeb can be accessed at: http://wodaklab.org/iRefWeb. Database URL: http://wodaklab.org/iRefWeb/

Effect of Voxel Size on 3D Micro-CT Analysis of Cortical Bone Porosity

This study examines the impact of voxel size on 3D micro-CT analysis of human cortical bone porosity. The study is based on computed microtomography scans of 10 human anterior femoral midshaft specimens acquired at 5, 10, and 15 μm voxel sizes. Artificial voxel sizes (10, 20, and 40 μm) were generated from the smallest scan voxel size (5 μm) in order to compare actual scanning with artificial degradation, a method employed in other similar studies. Canal volume fraction (CaV/TV), canal surface to volume ratio (CaS/CaV), mean canal diameter (CaDm), mean canal separation (CaSp), canal number (CaN), degree of anisotropy (DA), and canal connectivity density (CaConnD) were calculated from matching volumes of interest for all datasets. Qualitatively, the clarity of the actual scan datasets deteriorated rapidly as voxel size increased. In contrast, within the artificially generated datasets, the clarity of cortical pores was better maintained until the largest voxel size (40 μm). Mean absolute percent error values, correlation coefficients, and paired t-tests revealed a pattern of increasing, and generally significant, differences between the smallest and progressively larger voxel sizes (both scanned and artificial). Relative to the actual scans, however, the artificial datasets were less sensitive to changing voxel size. These findings indicated that subtle changes in voxel size, within the range examined, have a considerable effect on human cortical porosity structural parameters. Additionally, the use of artificially increased voxel sizes should be viewed with caution as they may not reflect what can actually be obtained by scanning.

Literature curation of protein interactions: measuring agreement across major public databases

Literature curation of protein interaction data faces a number of challenges. Although curators increasingly adhere to standard data representations, the data that various databases actually record from the same published information may differ significantly. Some of the reasons underlying these differences are well known, but their global impact on the interactions collectively curated by major public databases has not been evaluated. Here we quantify the agreement between curated interactions from 15 471 publications shared across nine major public databases. Results show that on average, two databases fully agree on 42% of the interactions and 62% of the proteins curated from the same publication. Furthermore, a sizable fraction of the measured differences can be attributed to divergent assignments of organism or splice isoforms, different organism focus and alternative representations of multi-protein complexes. Our findings highlight the impact of divergent curation policies across databases, and should be relevant to both curators and data consumers interested in analyzing protein-interaction data generated by the scientific community. Database URL: http://wodaklab.org/iRefWeb

Protein–protein interaction networks: the puzzling riches

While major progress has been achieved in the experimental techniques used for the detection of protein interactions and in the processing and analysis of the vast amount of data that they generate, we still do not understand why the set of identified interactions remains so highly dependent on the particular detection method. Here we present an overview of the major high-throughput experimental methods used to detect interactions and the datasets produced using these methods over the last 10 years. We discuss the challenges of assessing the quality of these datasets, and examine key factors that likely underlie the persistent poor overlap between the interactions detected by different methods. Lastly, we present a brief overview of the literature-curated protein interaction data stored in public databases, which are often relied upon for independent validation of newly derived interaction networks.

Multilocus loss of DNA methylation in individuals with mutations in the histone H3 Lysine 4 Demethylase KDM5C

BackgroundA number of neurodevelopmental syndromes are caused by mutations in genes encoding proteins that normally function in epigenetic regulation. Identification of epigenetic alterations occurring in these disorders could shed light on molecular pathways relevant to neurodevelopment.ResultsUsing a genome-wide approach, we identified genes with significant loss of DNA methylation in blood of males with intellectual disability and mutations in the X-linked KDM5C gene, encoding a histone H3 lysine 4 demethylase, in comparison to age/sex matched controls. Loss of DNA methylation in such individuals is consistent with known interactions between DNA methylation and H3 lysine 4 methylation. Further, loss of DNA methylation at the promoters of the three top candidate genes FBXL5, SCMH1, CACYBP was not observed in more than 900 population controls. We also found that DNA methylation at these three genes in blood correlated with dosage of KDM5C and its Y-linked homologue KDM5D. In addition, parallel sex-specific DNA methylation profiles in brain samples from control males and females were observed at FBXL5 and CACYBP.ConclusionsWe have, for the first time, identified epigenetic alterations in patient samples carrying a mutation in a gene involved in the regulation of histone modifications. These data support the concept that DNA methylation and H3 lysine 4 methylation are functionally interdependent. The data provide new insights into the molecular pathogenesis of intellectual disability. Further, our data suggest that some DNA methylation marks identified in blood can serve as biomarkers of epigenetic status in the brain.

Human-Chromatin-Related Protein Interactions Identify a Demethylase Complex Required for Chromosome Segregation

Chromatin regulation is driven by multicomponent protein complexes, which form functional modules. Deciphering the components of these modules and their interactions is central to understanding the molecular pathways these proteins are regulating, their functions, and their relation to both normal development and disease. We describe the use of affinity purifications of tagged human proteins coupled with mass spectrometry to generate a protein-protein interaction map encompassing known and predicted chromatin-related proteins. On the basis of 1,394 successful purifications of 293 proteins, we report a high-confidence (85% precision) network involving 11,464 protein-protein interactions among 1,738 different human proteins, grouped into 164 often overlapping protein complexes with a particular focus on the family of JmjC-containing lysine demethylases, their partners, and their roles in chromatin remodeling. We show that RCCD1 is a partner of histone H3K36 demethylase KDM8 and demonstrate that both are important for cell-cycle-regulated transcriptional repression in centromeric regions and accurate mitotic division.

Intercellular network structure and regulatory motifs in the human hematopoietic system

The hematopoietic system is a distributed tissue that consists of functionally distinct cell types continuously produced through hematopoietic stem cell (HSC) differentiation. Combining genomic and phenotypic data with high‐content experiments, we have built a directional cell–cell communication network between 12 cell types isolated from human umbilical cord blood. Network structure analysis revealed that ligand production is cell type dependent, whereas ligand binding is promiscuous. Consequently, additional control strategies such as cell frequency modulation and compartmentalization were needed to achieve specificity in HSC fate regulation. Incorporating the in vitro effects (quiescence, self‐renewal, proliferation, or differentiation) of 27 HSC binding ligands into the topology of the cell–cell communication network allowed coding of cell type‐dependent feedback regulation of HSC fate. Pathway enrichment analysis identified intracellular regulatory motifs enriched in these cell type‐ and ligand‐coupled responses. This study uncovers cellular mechanisms of hematopoietic cell feedback in HSC fate regulation, provides insight into the design principles of the human hematopoietic system, and serves as a foundation for the analysis of intercellular regulation in multicellular systems.

Interaction databases on the same page

391 to drive continued investment in their development. If healthcare payers fail to incentivize the sector and set overwhelming barriers to innovation, it will considerably hinder progress in personalized medicine. Test developers as well as regulatory authorities are accepting risks by developing, approving and championing molecular diagnostics; healthcare payers must now also share an element of this risk by adopting a positive stance toward coverage and reimbursement of molecular diagnostic technology.

The missing indels: an estimate of indel variation in a human genome and analysis of factors that impede detection

With the development of High-Throughput Sequencing (HTS) thousands of human genomes have now been sequenced. Whenever different studies analyze the same genome they usually agree on the amount of single-nucleotide polymorphisms, but differ dramatically on the number of insertion and deletion variants (indels). Furthermore, there is evidence that indels are often severely under-reported. In this manuscript we derive the total number of indel variants in a human genome by combining data from different sequencing technologies, while assessing the indel detection accuracy. Our estimate of approximately 1 million indels in a Yoruban genome is much higher than the results reported in several recent HTS studies. We identify two key sources of difficulties in indel detection: the insufficient coverage, read length or alignment quality; and the presence of repeats, including short interspersed elements and homopolymers/dimers. We quantify the effect of these factors on indel detection. The quality of sequencing data plays a major role in improving indel detection by HTS methods. However, many indels exist in long homopolymers and repeats, where their detection is severely impeded. The true number of indel events is likely even higher than our current estimates, and new techniques and technologies will be required to detect them.

NSD1 mutations generate a genome-wide DNA methylation signature

Sotos syndrome (SS) represents an important human model system for the study of epigenetic regulation; it is an overgrowth/intellectual disability syndrome caused by mutations in a histone methyltransferase, NSD1. As layered epigenetic modifications are often interdependent, we propose that pathogenic NSD1 mutations have a genome-wide impact on the most stable epigenetic mark, DNA methylation (DNAm). By interrogating DNAm in SS patients, we identify a genome-wide, highly significant NSD1+/−-specific signature that differentiates pathogenic NSD1 mutations from controls, benign NSD1 variants and the clinically overlapping Weaver syndrome. Validation studies of independent cohorts of SS and controls assigned 100% of these samples correctly. This highly specific and sensitive NSD1+/− signature encompasses genes that function in cellular morphogenesis and neuronal differentiation, reflecting cardinal features of the SS phenotype. The identification of SS-specific genome-wide DNAm alterations will facilitate both the elucidation of the molecular pathophysiology of SS and the development of improved diagnostic testing.