Biology subjects

Fu, D. Y.

Publications and source records attributed to Fu, D. Y..

4 recordsLinked to original sources

bioRxiv2019

Releasing a preprint is associated with more attention and citations

Preprints in biology are gaining popularity, but release of a preprint still precedes only a fraction of peer-reviewed publications. We examined whether having a preprint on bioRxiv was associated with metrics of the corresponding peer-reviewed article. We assembled a dataset of 74,239 articles, 5,405 of which had a preprint, published in 39 journals. Based on log-linear regression and random-effects meta-analysis, articles with a preprint had a 51% higher Altmetric Attention Score and 37% more citations compared to articles without one. These associations were independent of several other article- and author-level variables (e.g., scientific subfield and last author publication age) and unrelated to journal-level variables such as access model and Impact Factor. This observational study can help researchers and publishers make informed decisions about how to incorporate preprints into their work.

scientific communication and education↗

bioRxiv2019

Cox regression increases power to detect genotype-phenotype associations in genomic studies using the electronic health record

BackgroundThe growth of DNA biobanks linked to data from electronic health records (EHRs) has enabled the discovery of numerous associations between genomic variants and clinical phenotypes. Nonetheless, although clinical data are generally longitudinal, standard approaches for detecting genotype-phenotype associations in such linked data, notably logistic regression, do not naturally account for the times at which events occur. Here we explored the advantages of quantifying associations using Cox proportional hazards regression, which can account for the age at which a patient first visited the healthcare system (left truncation) and the age at which a patient either last visited the healthcare system or acquired a particular phenotype (right censoring).\n\nResultsUsing simulated data, we found that, compared to logistic regression, Cox regression had greater power at equivalent Type I error. We then scanned for genotype-phenotype associations using logistic regression and Cox regression on 50 phenotypes derived from the electronic health records of 49 792 genotyped individuals. In terms of effect sizes, the hazard ratios estimated by Cox regression were nearly identical to the odds ratios estimated by logistic regression. Consistent with the findings from our simulations, Cox regression had approximately 10% greater relative sensitivity for detecting known associations from the NHGRI-EBI GWAS Catalog.\n\nConclusionsAs longitudinal health-related data continue to grow, Cox regression may improve our ability to identify the genetic basis for a wide range of human phenotypes.

bioinformatics↗

bioRxiv2018

Simphony: simulating large-scale, rhythmic data

Simulated data are invaluable for assessing a computational method's ability to distinguish signal from noise. Although many biological systems show rhythmicity, there is no general-purpose tool to simulate large-scale, rhythmic data. Here we present Simphony, an R package for simulating data from experiments in which the abundances of rhythmic and non-rhythmic features (e.g., genes) are measured at multiple time points in multiple conditions. Simphony has parameters for specifying experimental design and each feature's rhythmic properties (e.g., shape, amplitude, and phase). In addition, Simphony can sample measurements from Gaussian and negative binomial distributions, the latter of which approximates read counts from next-generation sequencing data. We show an example of using Simphony to benchmark a method for detecting rhythms. Our results suggest that Simphony can aid experimental design and computational method development. Simphony is thoroughly documented and freely available at https://github.com/hugheylab/simphony.

bioinformatics↗

bioRxiv2018

Protein-Ligand Docking with Protein-based and Ligand-based Structure Activity Relationships

Protein-small molecule docking programs predict the interaction interface and energy between a given protein target and a small molecule ligand. The accuracy of docking predictions generally improve with the guidance of experimentally derived restraints. One available source of such restraints is structure-activity relationships (SARs). SARs provide information on changes in binding affinity or biological response corresponding to a chemical change in the protein and/or ligand. These chemical changes frequently refer to amino acid mutations on the protein side and functional group modifications on the ligand side. Theoretically, predicted interaction energies should correlate with SARs though in practice, this is challenging due to the difficulties in scoring protein-ligand interactions. We have previously developed RosettaLigandEnsemble (RLE), a protein-ligand docking method that simultaneously docks a congeneric ligand series to a single protein target. RLE is capable of identifying native-like binding modes for a ligand series that match the available ligand SARs. This work in progress reports on the extension of RLE to factor in SARs derived from protein mutagenesis data. The new method, ProteinLigEnsemble (PLE), is also part of the Rosetta Biomolecular Modeling Suite available at https://www.rosettacommons.org/. We have also developed protein ensemble docking features that allow for docking or screening against multiple receptor variants at the same time. We have included a proof of concept study and a tutorial for interested users.

biophysics↗