Polygenic scores for major depressive disorder and depressive symptoms predict response to lithium in patients with bipolar disorder, bioRxiv, 2018-10-22

AbstractBackgroundLithium is a first-line medication for bipolar disorder (BD), but only ~30% of patients respond optimally to the drug. Since genetic factors are known to mediate lithium treatment response, we hypothesized whether polygenic susceptibility to the spectrum of depression traits is associated with treatment outcomes in patients with BD. In addition, we explored the potential molecular underpinnings of this relationship.MethodsWeighted polygenic scores (PGSs) were computed for major depressive disorder (MDD) and depressive symptoms (DS) in BD patients from the Consortium on Lithium Genetics (ConLi+Gen; n=2,586) who received lithium treatment. Lithium treatment outcome was assessed using the ALDA scale. Summary statistics from genome-wide association studies (GWAS) in MDD (130,664 cases and 330,470 controls) and DS (n=161,460) were used for PGS weighting. Associations between PGSs of depression traits and lithium treatment response were assessed by binary logistic regression. We also performed a cross-trait meta-GWAS, followed by Ingenuity® Pathway Analysis.OutcomesBD patients with a low polygenic load for depressive traits were more likely to respond well to lithium, compared to patients with high polygenic load (MDD OR =1.64 [95%CI 1.26-2.15], lowest vs highest PGS quartiles; DS OR=1.53 [95%CI 1.18-2.00]). Associations were significant for type 1, but not type 2 BD. Cross-trait GWAS and functional characterization implicated voltage-gated potassium channels, insulin-related pathways, mitogen-activated protein-kinase (MAPK) signaling, and miRNA expression.InterpretationGenetic loading to depression traits in BD patients lower their odds of responding optimally to lithium. Our findings support the emerging concept of a lithium-responsive biotype in BD.FundingSee attached details

The population history of northeastern Siberia since the Pleistocene, bioRxiv, 2018-10-22

ABSTRACTFar northeastern Siberia has been occupied by humans for more than 40 thousand years. Yet, owing to a scarcity of early archaeological sites and human remains, its population history and relationship to ancient and modern populations across Eurasia and the Americas are poorly understood. Here, we analyze 34 ancient genome sequences, including two from fragmented milk teeth found at the ~31.6 thousand-year-old (kya) Yana RHS site, the earliest and northernmost Pleistocene human remains found. These genomes reveal complex patterns of past population admixture and replacement events throughout northeastern Siberia, with evidence for at least three large-scale human migrations into the region. The first inhabitants, a previously unknown population of “Ancient North Siberians” (ANS), represented by Yana RHS, diverged ~38 kya from Western Eurasians, soon after the latter split from East Asians. Between 20 and 11 kya, the ANS population was largely replaced by peoples with ancestry related to present-day East Asians, giving rise to ancestral Native Americans and “Ancient Paleosiberians” (AP), represented by a 9.8 kya skeleton from Kolyma River. AP are closely related to the Siberian ancestors of Native Americans, and ancestral to contemporary communities such as Koryaks and Itelmen. Paleoclimatic modelling shows evidence for a refuge during the last glacial maximum (LGM) in southeastern Beringia, suggesting Beringia as a possible location for the admixture forming both ancestral Native Americans and AP. Between 11 and 4 kya, AP were in turn largely replaced by another group of peoples with ancestry from East Asia, the “Neosiberians” from which many contemporary Siberians derive. We detect gene flow events in both directions across the Bering Strait during this time, influencing the genetic composition of Inuit, as well as Na Dene-speaking Northern Native Americans, whose Siberian-related ancestry components is closely related to AP. Our analyses reveal that the population history of northeastern Siberia was highly dynamic throughout the Late Pleistocene and Holocene. The pattern observed in northeastern Siberia, with earlier, once widespread populations being replaced by distinct peoples, seems to have taken place across northern Eurasia, as far west as Scandinavia.

Bridging the divide bacteria synthesizing archaeal membrane lipids, bioRxiv, 2018-10-19

Archaea synthesize membranes of isoprenoid lipids that are ether-linked to glycerol, while BacteriaEukarya produce membranes consisting of ester-bound fatty acids. This dichotomy in membrane lipid composition or ‘lipid divide’ is believed to have arisen after the Last Universal Common Ancestor (LUCA). A leading hypothesis is that LUCA possessed a ‘mixed heterochiral archaealbacterial membrane’, however no natural microbial representatives supporting this scenario have been shown to exist today. Here, we demonstrate that bacteria of the Fibrobacteres-Chlorobi-Bacteroidetes (FCB) group superphylum and related candidate phyla encode a complete pathway for archaeal membrane lipid biosynthesis in addition to the bacterial fatty acid membrane pathway. Key genes were expressed in the environment and their recombinant expression in E. coli resulted in the formation of a ‘mixed archaealbacterial membrane’. Our results support the existence of ‘mixed membranes’ in natural environments and their stability over large evolutionary timescales, thereby bridging a once-thought fundamental divide in biology.

Explaining heritable variance in human character, bioRxiv, 2018-10-19

A stable, long-term cortical signature underlying consistent behavior, bioRxiv, 2018-10-18

AbstractAnimals readily execute learned motor behaviors in a consistent manner over long periods of time, yet similarly stable neural correlates remained elusive up to now. How does the cortex achieve this stable control? Using the sensorimotor system as a model of cortical processing, we investigated the hypothesis that the dynamics of neural latent activity, which capture the dominant co-variation patterns within the neural population, are preserved across time. We recorded from populations of neurons in premotor, primary motor, and somatosensory cortices for up to two years as monkeys performed a reaching task. Intriguingly, despite steady turnover in the recorded neurons, the low-dimensional latent dynamics remained stable. Such stability allowed reliable decoding of behavioral features for the entire timespan, while fixed decoders based on the recorded neural activity degraded substantially. We posit that latent cortical dynamics within the manifold are the fundamental and stable building blocks underlying consistent behavioral execution.

Mutation detection in thousands of acute myeloid leukemia cells using single cell RNA-sequencing, bioRxiv, 2018-10-18

AbstractVirtually all tumors are genetically heterogeneous, containing subclonal populations of cells that are defined by distinct mutations1. Subclones can have unique phenotypes that influence disease progression2, but these phenotypes are difficult to characterize subclones usually cannot be physically purified, and bulk gene expression measurements obscure interclonal differences. Single-cell RNA-sequencing has revealed transcriptional heterogeneity within a variety of tumor types, but it is unclear how this expression heterogeneity relates to subclonal genetic events – for example, whether particular expression clusters correspond to mutationally defined subclones3,4,5,6-9. To address this question, we developed an approach that integrates enhanced whole genome sequencing (eWGS) with the 10x Genomics Chromium Single Cell 5’ Gene Expression workflow (scRNA-seq) to directly link expressed mutations with transcriptional profiles at single cell resolution. Using bone marrow samples from five cases of primary human Acute Myeloid Leukemia (AML), we generated WGS and scRNA-seq data for each case. Duplicate single cell libraries representing a median of 20,474 cells per case were generated from the bone marrow of each patient. Although the libraries were 5’ biased, we detected expressed mutations in cDNAs at distances up to 10 kbp from the 5’ ends of well-expressed genes, allowing us to identify hundreds to thousands of cells with AML-specific somatic mutations in every case. This data made it possible to distinguish AML cells (including normal-karyotype AML cells) from surrounding normal cells, to study tumor differentiation and intratumoral expression heterogeneity, to identify expression signatures associated with subclonal mutations, and to find cell surface markers that could be used to purify subclones for further study. The data also revealed transcriptional heterogeneity that occurred independently of subclonal mutations, suggesting that additional factors drive epigenetic heterogeneity. This integrative approach for connecting genotype to phenotype in AML cells is broadly applicable for analysis of any sample that is phenotypically and genetically heterogeneous.