mosGraphGen: a novel tool to generate multi-omic signaling graphs to facilitate integrative and interpretable graph AI model development.

Multi-omic data, i.e., genomics, epigenomics, transcriptomics, proteomics, characterize cellular complex signaling systems from multi-level and multi-view and provide a holistic view of complex cellular signaling pathways. However, it remains challenging to integrate and interpret multi-omics data. Graph neural network (GNN) AI models have been widely used to analyze graph-structure datasets and are ideal for integrative multi-omics data analysis because they can naturally integrate and represent multi-omics data as a biologically meaningful multi-level signaling graph and interpret multi-omics data by node and edge ranking analysis for signaling flow/cascade inference. However, it is non-trivial for graph-AI model developers to pre-analyze multi-omics data and convert them into graph-structure data for individual samples, which can be directly fed into graph-AI models. To resolve this challenge, we developed mosGraphGen (multi-omics signaling graph generator), a novel computational tool that generates multi-omics signaling graphs of individual samples by mapping the multi-omics data onto a biologically meaningful multi-level background signaling network. With mosGraphGen, AI model developers can directly apply and evaluate their models using these mos-graphs. We evaluated the mosGraphGen using both multi-omics datasets of cancer and Alzheimer's disease (AD) samples. The code of mosGraphGen is open-source and publicly available via GitHub: https://github.com/Multi-OmicGraphBuilder/mosGraphGen.

Biologic characterization of ABCA3 variants in lung tissue from infants and children with ABCA3 deficiency.

ABCA3 is a phospholipid transporter protein required for surfactant assembly in lamellar bodies of alveolar type II cells. Biallelic pathogenic ABCA3 variants cause severe neonatal respiratory distress syndrome or childhood interstitial lung disease. However, ABCA3 genotype alone does not explain the diversity in disease presentation, severity, and progression. Additionally, monoallelic ABCA3 variants have been reported in infants and children with ABCA3-deficient phenotypes. The effects of most ABCA3 variants identified in patients have not been characterized at the RNA level. ABCA3 allele-specific expression occurs in some cell types due to epigenetic regulation. We obtained lung tissue at transplant or autopsy from 16 infants and children with ABCA3 deficiency due to compound heterozygous ABCA3 variants for biologic characterization of the predicted effects of ABCA3 variants at the RNA level and determination of ABCA3 allele expression. We extracted DNA and RNA from frozen lung tissue and reverse-transcribed cDNA from mRNA. We performed Sanger sequencing to assess allele-specific expression by comparing the heights of variant nucleotide peaks in amplicons from genomic DNA and cDNA. We found similar genomic and cDNA variant nucleotide peak heights and no evidence of allele-specific expression among explant or autopsy samples with biallelic missense ABCA3 variants (n = 6). We observed allele-specific expression of missense alleles in trans with frameshift (n = 4) or nonsense (n = 1) variants, attributable to nonsense-mediated decay. The missense variant c.53 A > G;p.Gln18Arg, located near an exon-intron junction, encoded abnormal splicing with skipping of exon 4. Biologic characterization of ABCA3 variants can inform discovery of variant-specific disease mechanisms.

Functional characterization of four ATP-binding cassette transporter A3 gene (ABCA3) variants.

ABCA3 transports phospholipids across lamellar body membranes in pulmonary alveolar type II cells and is required for surfactant assembly. Rare, biallelic, pathogenic ABCA3 variants result in lethal neonatal respiratory distress syndrome and childhood interstitial lung disease. Qualitative functional characterization of ABCA3 missense variants suggests two pathogenic classes: disrupted intracellular trafficking (type I mutant) or impaired ATPase-mediated phospholipid transport into the lamellar bodies (type II mutant). We qualitatively compared wild-type (WT-ABCA3) with four uncharacterized ABCA3 variants (c.418A>C;p.Asn140His, c.3609_3611delCTT;p.Phe1203del, c.3784A>G;p.Ser1262Gly, and c.4195G>A;p.Val1399Met) in A549 cells using protein processing, colocalization with intracellular organelles, lamellar body ultrastructure, and ATPase activity. We quantitatively measured lamellar body-like vesicle diameter and intracellular ABCA3 trafficking using fluorescence-based colocalization. Three ABCA3 variants (p.Asn140His, p.Ser1262Gly, and p.Val1399Met) were processed and trafficked normally and demonstrated well-organized lamellar body-like vesicles, but had reduced ATPase activity consistent with type II mutants. P.Phe1203del was processed normally, had reduced ATPase activity, and well-organized lamellar body-like vesicles, but quantitatively colocalized with both endoplasmic reticulum and lysosomal markers, an intermediate phenotype suggesting disruption of both intracellular trafficking and phospholipid transport. All ABCA3 mutants demonstrated mean vesicle diameters smaller than WT-ABCA3. Qualitative and quantitative functional characterization of ABCA3 variants informs mechanisms of pathogenicity.