LRLoop: a method to predict feedback loops in cell–cell communication

Abstract

Motivation Intercellular communication (i.e. cell-cell communication) plays an essential role in multicellular organisms coordinating various biological processes. Previous studies discovered that feedback loops between two cell types are a widespread and vital signaling motif regulating development, regeneration, and cancer progression. While many computational methods have been developed to predict cell-cell communication based on gene expression datasets, these methods often predict one-directional ligand-receptor interactions from sender to receiver cells and are not suitable to identify feedback loops. Results Here we describe LRLoop, a new method for analyzing cell-cell communication based on bi-directional ligand-receptor interactions, where two pairs of ligand-receptor interactions are identified that are responsive to each other, and thereby form a closed feedback loop. We first assessed LRLoop using bulk datasets and found our method significantly reduces the false positive rate seen with existing methods. Furthermore, we developed a new strategy to assess the performance of these methods in single-cell datasets. We used the between-tissue interactions as an indicator of potential false-positive prediction and found that LRLoop produced a lower fraction of between-tissue interactions than traditional methods. Finally, we applied LRLoop to the single-cell datasets obtained from retinal development. We discovered many new bi-directional ligand-receptor interactions among individual cell types that potentially control proliferation, neurogenesis, and/or cell fate specification. Availability An R package is available at https://github.com/Pinlyu3/LRLoop. The source code can be found at figureshare (https://doi.org/10.6084/m9.figshare.20126138.v1). The datasets can be found at figureshare (https://doi.org/10.6084/m9.figshare.20126021.v1). Supplementary information Supplementary data are available at Bioinformatics online.

Microglia–Neutrophil Interactions Drive Dry AMD-like Pathology in a Mouse Model

“…Further, in-depth analysis revealed several differentially expressed genes in the RPE and microglial cells ( Figure 3 b,c). To pinpoint the signaling molecules associated with RPE–microglia interaction in the disease state of this mouse model, we performed a novel ligand-receptor (LR) loop bioinformatic analysis [ 25 ] that showed several interactive partners linking RPE (Ligand; L) and microglia (Receptor; R) in 15-month-old Cryba1 fl/fl and Cryba1 cKO ( Figure 3 d). The LR loop interaction between RPE and microglia revealed a notable increase in the interactions between cognate ligands from RPE cells (such as CXCL12 and Col4a3) and their respective cell surface receptors such as Itgb1 and CD47 on microglia.…”

“…Currently, the extent that microglia activate neutrophils or how these immune cells interact with each other during AMD progression in the SRS remains unclear. To understand the dynamic interactions between these immune cells, we used LR loop bioinformatic analysis [ 25 ], which showed several interactive partners linking microglia and neutrophils and several differentially expressed microglial ( Figure 3 c) and neutrophil ( Figure 5 a) genes. Interestingly, we found that both CD14 (microglia)/integrin β1 (Itgβ1) (neutrophil) and CD14/integrin α4 (Itgα4) interactions were upregulated in Cryba1 cKO retina ( Figure 5 a,b).…”

“…In other diseases such as Alzheimer’s disease, stroke, and intracerebral hemorrhage, it has been previously documented that activated microglia facilitate neutrophil survival, activation, and migration leading to tissue damage [ 45 , 46 , 47 ]. To assess the role of microglia/neutrophil interaction in AMD progression, we performed scRNASeq analysis on cells from the SRS isolated from Cryba1 cKO mice employing a novel bioinformatic tool, LR-loop [ 25 ]. This identified several interacting partners linking the two cell types, suggesting that microglia–neutrophil interaction is an important factor in the AMD-like pathology in this mouse model.…”

“…For the analysis of ligand–receptor interactions between cells, the default built-in ligand-receptor interaction database, intracellular signaling database and gene regulatory networks in the package LRLoop were used [ 25 ]. The ligand–receptor pairs and intracellular signaling interactions with a detection rate less than 2.5% in the corresponding cell types were removed [ 25 ].…”

“…For the analysis of ligand–receptor interactions between cells, the default built-in ligand-receptor interaction database, intracellular signaling database and gene regulatory networks in the package LRLoop were used [ 25 ]. The ligand–receptor pairs and intracellular signaling interactions with a detection rate less than 2.5% in the corresponding cell types were removed [ 25 ]. Analysis of the remaining candidate ligand–receptor pairs between each pair of cell types of interest in each sample was performed using the LRLoop package with its standard pipeline and default parameters [ 23 ].…”