We present a systematic evaluation of state-of-the-art algorithms for inferring gene regulatory networks (GRNs) from single-cell transcriptional data. As the ground truth for assessing accuracy, we use synthetic networks with predictable trajectories, literature-curated Boolean models, and diverse transcriptional regulatory networks. We develop a strategy to simulate single-cell transcriptional data from synthetic and Boolean networks that avoids pitfalls of previously-used methods. Furthermore, we collect networks from multiple experimental single-cell RNA-seq datasets. We develop an evaluation framework called BEELINE. We find that the AUPRC and early precision of the algorithms are moderate. The methods are better in recovering interactions in synthetic networks than Boolean models. The algorithms with the best early precision values for Boolean models also perform well on experimental datasets. Techniques that do not require pseudotime-ordered cells are generally more accurate. Based on these results, we present recommendations to end users. BEELINE will aid the development of GRN inference algorithms.Single-cell RNA-sequencing technology has made it possible to trace cellular lineages during differentiation and to identify new cell types 1,2 . A central question that arises now is whether we can discover the gene regulatory networks (GRNs) that control cellular differentiation and drive transitions from one cell type to another. In such a GRN, each edge connects a transcription factor (TF) to a gene it regulates. Ideally, the edge is directed from the TF to the target gene, represents direct rather than indirect regulation, and corresponds to activation or inhibition.Single-cell expression data are especially promising for computing GRNs because, unlike bulk transcriptomic data, they do not obscure biological signals by averaging over all the cells in a sample. However, these data have features that pose significant difficulties, e.g., Users may view, print, copy, and download text and data-mine the content in such documents, for the purposes of academic research, subject always to the full Conditions of use:

We present a comprehensive evaluation of state-of-the-art algorithms for inferring gene regulatory networks (GRNs) from single-cell gene expression data. Our contributions include a comprehensive evaluation pipeline based on simulated data from "toy", artificial networks with predictable cellular trajectories and on simulated data from carefully-curated Boolean models. We develop a strategy to simulate these two types of data that avoids the pitfalls of existing strategies that have been used to mimic bulk transcriptional data. We found that the accuracy of the algorithms measured in terms of AUROC and AUPRC was moderate, by and large, although the methods were better in recovering interactions in the artificial networks than the Boolean models. Techniques that did not require pseudotime-ordered cells were more accurate, in general. There were an excess of feedforward loops in predicted networks than in the Boolean models. The observation that the endpoints of many false positive edges were connected by paths of length two in the Boolean models suggested that indirect effects may be predominant in the outputs of these algorithms. The outputs of the methods were quite inconsistent with each other, indicating that combining these approaches using ensembles is likely to be challenging. We present recommendations on how to create simulated gene expression datasets for testing GRN inference algorithms. We suggest that new ideas for avoiding the prediction of indirect interactions appear to be necessary to improve the accuracy of GRN inference algorithms for single cell gene expression data. Simulated data from synthetic networks Simulated data from curated models LEAP PIDC GRN inference methods Predicted networks SCODE LEAP PIDC SCODE Run algorithms Parameter search Software run time Evaluate network motifs ROC Early Precision Stability of inferred networks

Kisspeptins (Kp), peptide products of the Kisspeptin-1 (KISS1) gene are endogenous ligands for a G protein-coupled receptor 54 (GPR54). Previous findings have shown that KISS1 acts as a metastasis suppressor in numerous cancers in humans. However, recent studies have demonstrated that an increase in KISS1 and GPR54 expression in human breast tumors correlates with higher tumor grade and metastatic potential. At present, whether or not Kp signaling promotes breast cancer cell invasiveness, required for metastasis and the underlying mechanisms, is unknown. We have found that kisspeptin-10 (Kp-10), the most potent Kp, stimulates the invasion of human breast cancer MDA-MB-231 and Hs578T cells using Matrigel-coated Transwell chamber assays and induces the formation of invasive stellate structures in three-dimensional invasion assays. Furthermore, Kp-10 stimulated an increase in matrix metalloprotease (MMP)-9 activity. We also found that Kp-10 induced the transactivation of epidermal growth factor receptor (EGFR). Knockdown of the GPCR scaffolding protein, β-arrestin 2, inhibited Kp-10-induced EGFR transactivation as well as Kp-10 induced invasion of breast cancer cells via modulation of MMP-9 secretion and activity. Finally, we found that the two receptors associate with each other under basal conditions, and FRET analysis revealed that GPR54 interacts directly with EGFR. The stability of the receptor complex formation was increased upon treatment of cells by Kp-10. Taken together, our findings suggest a novel mechanism by which Kp signaling via GPR54 stimulates breast cancer cell invasiveness.

PathLinker is a graph-theoretic algorithm for reconstructing the interactions in a signaling pathway of interest. It efficiently computes multiple short paths within a background protein interaction network from the receptors to transcription factors (TFs) in a pathway. We originally developed PathLinker to complement manual curation of signaling pathways, which is slow and painstaking. The method can be used in general to connect any set of sources to any set of targets in an interaction network. The app presented here makes the PathLinker functionality available to Cytoscape users. We present an example where we used PathLinker to compute and analyze the network of interactions connecting proteins that are perturbed by the drug lovastatin.

Advances in the field of goal-directed molecular optimization offer the promise of finding feasible candidates for even the most challenging molecular design applications. One example of a fundamental design challenge is the search for novel stable radical scaffolds for an aqueous redox flow battery that simultaneously satisfy redox requirements at the anode and cathode, as relatively few stable organic radicals are known to exist. To meet this challenge, we develop a new open-source molecular optimization framework based on AlphaZero coupled with a fast, machine-learning-derived surrogate objective trained with nearly 100,000 quantum chemistry simulations. The objective function comprises two graph neural networks: one that predicts adiabatic oxidation and reduction potentials and a second that predicts electron density and local three-dimensional environment, previously shown to be correlated with radical persistence and stability. With no hard-coded knowledge of organic chemistry, the reinforcement learning agent finds molecule candidates that satisfy a precise combination of redox, stability and synthesizability requirements defined at the quantum chemistry level, many of which have reasonable predicted retrosynthetic pathways. The optimized molecules show that alternative stable radical scaffolds may offer a unique profile of stability and redox potentials to enable low-cost symmetric aqueous redox flow batteries.