Krinsman, W. E. | Search bioRxiv

bioRxiv2022

Models of Throughput for Multi-Cell, Multi-Type Droplet Microfluidics

New experimental platforms encapsulate multiple cells per microfluidic droplet, with each cell belonging to one of multiple possible types. The motivating example comes from microbial ecology, where we want to observe the interactions of microbial strains. Because droplets are formed randomly, we want to accurately predict the data throughput, the numbers of droplets containing desired combinations of cell types. Herein I identify the default statistical model for predicting the data throughput of multi-cell, multi-type droplet microfluidics experiments, which fits to cell type count data. I explain the assumptions behind this model and issues that in practice may cause these assumptions to fail. One such issue, "compositional heterogeneity", is unique to multi-type experiments. I show how to modify the default statistical model to describe the consequences of these issues, without needing to mechanistically model their causes. In practice, only two of these issues may substantially change the data throughput predictions. The changes depend on both (1) which combination of these issues are present, and (2) the precise definition of data throughput. Finally, I show that for a given experimental platform one can estimate the severity of these two issues, enabling more accurate data throughput predictions that account for these two issues.

bioinformatics↗

bioRxiv2022

Extending Comparison Methods for Unsigned Networks to Signed Networks

We can allow the edges of networks to have both negative and positive weights. For example, signed networks can describe the interactions of microbes. To evaluate the performance of estimators for signed networks, we need quantitative comparison methods for signed networks. Finding such comparison methods is done most easily by extending a comparison method for unsigned networks. Almost all methods reported in the literature for quantitatively comparing networks implicitly assume that edge weights are non-negative. Naive attempts to modify these methods to be applicable to signed networks can lead to nonsensical conclusions. Herein I identify requirements that should be satisfied by reasonable methods for comparing signed networks, most importantly the "double penalization principle". I extend several comparison methods for unsigned networks while satisfying these requirements. Finally, I give examples where these extensions behave reasonably but naive extensions do not.

bioinformatics↗