LRRK2 mediates tubulation and vesicle sorting from membrane damaged lysosomes, bioRxiv, 2020-01-25
ABSTRACTMutations in the leucine rich repeat kinase 2 (LRRK2) gene are a cause of familial and sporadic Parkinson’s disease (PD). Nonetheless, the biological functions of LRRK2 remain incompletely understood. Here, we observed that LRRK2 is recruited to lysosomes that have a ruptured membrane. Using unbiased proteomics, we observed that LRRK2 is able to recruit the motor adaptor protein JIP4 to permeabilized lysosomes in a kinase-dependent manner through the phosphorylation of RAB35 and RAB10. Super-resolution live cell imaging microscopy and FIB-SEM revealed that once at the lysosomal membrane, JIP4 promotes the formation of LAMP1-negative lysosomal tubules that release membranous content from ruptured lysosomes. Released vesicular structures are able to interact with other lysosomes. Thus, we described a new process that uses lysosomal tubulation to release vesicular structures from permeabilized lysosomes. LRRK2 orchestrates this process that we name LYTL (LYsosomal Tubulationsorting driven by LRRK2) that, given the central role of the lysosome in PD, is likely to be disease relevant.
biorxiv cell-biology 0-100-users 2020Low-N protein engineering with data-efficient deep learning, bioRxiv, 2020-01-24
AbstractProtein engineering has enormous academic and industrial potential. However, it is limited by the lack of experimental assays that are consistent with the design goal and sufficiently high-throughput to find rare, enhanced variants. Here we introduce a machine learning-guided paradigm that can use as few as 24 functionally assayed mutant sequences to build an accurate virtual fitness landscape and screen ten million sequences via in silico directed evolution. As demonstrated in two highly dissimilar proteins, avGFP and TEM-1 β-lactamase, top candidates from a single round are diverse and as active as engineered mutants obtained from previous multi-year, high-throughput efforts. Because it distills information from both global and local sequence landscapes, our model approximates protein function even before receiving experimental data, and generalizes from only single mutations to propose high-functioning epistatically non-trivial designs. With reproducible >500% improvements in activity from a single assay in a 96-well plate, we demonstrate the strongest generalization observed in machine-learning guided protein design to date. Taken together, our approach enables efficient use of resource intensive high-fidelity assays without sacrificing throughput. By encouraging alignment with endpoint objectives, low-N design will accelerate engineered proteins into the fermenter, field, and clinic.
biorxiv synthetic-biology 0-100-users 2020BlastFrost Fast querying of 100,000s of bacterial genomes in Bifrost graphs, bioRxiv, 2020-01-23
AbstractBlastFrost is a highly efficient method for querying 100,000s of genome assemblies. It builds on Bifrost, a recently developed dynamic data structure for compacted and colored de Bruijn graphs from bacterial genomes. BlastFrost queries a Bifrost data structure for sequences of interest, and extracts local subgraphs, thereby enabling the efficient identification of the presence or absence of individual genes or single nucleotide sequence variants. Here we describe the algorithms and implementation of BlastFrost. We also present two exemplar practical applications. In the first, we determined the presence of the individual genes within the SPI-2 Salmonella pathogenicity island within a collection of 926 representative genomes in minutes. In the second application, we determined the existence of known single nucleotide polymorphisms associated with fluoroquinolone resistance in the genes gyrA, gyrB and parE among 190, 209 Salmonella genomes. BlastFrost is available for download at <jatsext-link xmlnsxlink=httpwww.w3.org1999xlink ext-link-type=uri xlinkhref=httpsgithub.comnluhmannBlastFrost>httpsgithub.comnluhmannBlastFrost<jatsext-link>.
biorxiv bioinformatics 0-100-users 2020Human immune system variation during one year, bioRxiv, 2020-01-23
SUMMARYThe human immune system varies extensively between individuals, but variation within individuals over time has not been well characterized. Systems-level analyses allow for simultaneous quantification of many interacting immune system components, and the inference of global regulatory principles. Here we present a longitudinal, systems-level analysis in 99 healthy adults, 50 to 65 years of age and sampled every 3rd month during one year. We describe the structure of inter-individual variation and characterize extreme phenotypes along a principal curve. From coordinated measurement fluctuations, we infer relationships between 115 immune cell populations and 750 plasma proteins constituting the blood immune system. While most individuals have stable immune systems, the degree of longitudinal variability is an individual feature. The most variable individuals, in the absence of overt infections, exhibited markers of poor metabolic health suggestive of a functional link between metabolic and immunologic homeostatic regulation.HIGHLIGHTSLongitudinal variation in immune cell composition during one yearInter-individual variation can be described along a principal curveImmune cell and protein relationships are inferredVariability over time is an individual feature correlating with markers of poor metabolic health
biorxiv immunology 0-100-users 2020Post-prediction Inference, bioRxiv, 2020-01-23
AbstractMany modern problems in medicine and public health leverage machine learning methods to predict outcomes based on observable covariates [1, 2, 3, 4]. In an increasingly wide array of settings, these predicted outcomes are used in subsequent statistical analysis, often without accounting for the distinction between observed and predicted outcomes [1, 5, 6, 7, 8, 9]. We call inference with predicted outcomes post-prediction inference. In this paper, we develop methods for correcting statistical inference using outcomes predicted with an arbitrary machine learning method. Rather than trying to derive the correction from the first principles for each machine learning tool, we make the observation that there is typically a low-dimensional and easily modeled representation of the relationship between the observed and predicted outcomes. We build an approach for the post-prediction inference that naturally fits into the standard machine learning framework. We estimate the relationship between the observed and predicted outcomes on the testing set and use that model to correct inference on the validation set and subsequent statistical models. We show our postpi approach can correct bias and improve variance estimation (and thus subsequent statistical inference) with predicted outcome data. To show the broad range of applicability of our approach, we show postpi can improve inference in two totally distinct fields modeling predicted phenotypes in repurposed gene expression data [10] and modeling predicted causes of death in verbal autopsy data [11]. We have made our method available through an open-source R package [<jatsext-link xmlnsxlink=httpwww.w3.org1999xlink ext-link-type=uri xlinkhref=httpsgithub.comSiruoWangpostpi>httpsgithub.comSiruoWangpostpi<jatsext-link>]
biorxiv bioinformatics 0-100-users 2020High-resolution cryo-EM using beam-image shift at 200 keV, bioRxiv, 2020-01-22
ABSTRACTRecent advances in single-particle cryo-electron microscopy (cryo-EM) data collection utilizes beam-image shift to improve throughput. Despite implementation on well-aligned 300 keV cryo-EM instruments, it remains unknown how well beam-image shift data collection affects data quality on 200 keV instruments and whether any aberrations can be computationally corrected. To test this, we collected and analyzed a cryo-EM dataset of aldolase at 200 keV using beam-image shift. This analysis shows that beam tilt on the instrument initially limited the resolution of aldolase to 5.6Å. After iterative rounds of aberration correction and particle polishing in RELION, we were able to obtain a 2.8Å structure. This analysis indicates that software correction of microscope misalignment can provide a dramatic improvement in resolution.
biorxiv biophysics 0-100-users 2020