Machine learning-enhanced noninvasive prenatal testing of monogenic disorders.

OBJECTIVE: Single-nucleotide variants (SNVs) are of great significance in prenatal diagnosis as they are the leading cause of inherited single-gene disorders (SGDs). Identifying SNVs in a non-invasive prenatal screening (NIPS) scenario is particularly challenging for maternally inherited SNVs. We present an improved method to predict inherited SNVs from maternal or paternal origin in a genome-wide manner. METHODS: We performed SNV-NIPS based on the combination of fragments of cell free DNA (cfDNA) features, Bayesian inference and a machine-learning (ML) prediction refinement step using random forest (RF) classifiers trained on millions of non-pathogenic variants. We next evaluate the real-world performance of our refined method in a clinical setting by testing our models on 16 families with singleton pregnancies and varying fetal fraction (FF) levels, and validate the results over millions of inherited variants in each fetus. RESULTS: The average area under the ROC curve (AUC) values are 0.996 over all families for paternally inherited variants, 0.81 for the challenging maternally inherited variants, 0.86 for homozygous biallelic variants and 0.95 for compound heterozygous variants. Discriminative AUCs were achieved even in families with a low FF. We further investigate the performance of our method in correctly predicting SNVs in coding regions of clinically relevant genes and demonstrate significantly improved AUCs in these regions. Finally, we focus on the pathogenic variants in our cohort and show that our method correctly predicts if the fetus is unaffected or affected in all (10/10, 100%) of the families containing a pathogenic SNV. CONCLUSIONS: Overall, we demonstrate our ability to perform genome-wide NIPS for maternal and homozygous biallelic variants and showcase the utility of our method in a clinical setting.

Essential transcription factors for induced neuron differentiation.

Neurogenins are proneural transcription factors required to specify neuronal identity. Their overexpression in human pluripotent stem cells rapidly produces cortical-like neurons with spiking activity and, because of this, they have been widely adopted for human neuron disease models. However, we do not fully understand the key downstream regulatory effectors responsible for driving neural differentiation. Here, using inducible expression of NEUROG1 and NEUROG2, we identify transcription factors (TFs) required for directed neuronal differentiation by combining expression and chromatin accessibility analyses with a pooled in vitro CRISPR-Cas9 screen targeting all ~1900 TFs in the human genome. The loss of one of these essential TFs (ZBTB18) yields few MAP2-positive neurons. Differentiated ZBTB18-null cells have radically altered gene expression, leading to cytoskeletal defects and stunted neurites and spines. In addition to identifying key downstream TFs for neuronal differentiation, our work develops an integrative multi-omics and TFome-wide perturbation platform to rapidly characterize essential TFs for the differentiation of any human cell type.

Temperature-dependent RNA editing in octopus extensively recodes the neural proteome.

In poikilotherms, temperature changes challenge the integration of physiological function. Within the complex nervous systems of the behaviorally sophisticated coleoid cephalopods, these problems are substantial. RNA editing by adenosine deamination is a well-positioned mechanism for environmental acclimation. We report that the neural proteome of Octopus bimaculoides undergoes massive reconfigurations via RNA editing following a temperature challenge. Over 13,000 codons are affected, and many alter proteins that are vital for neural processes. For two highly temperature-sensitive examples, recoding tunes protein function. For synaptotagmin, a key component of Ca2+-dependent neurotransmitter release, crystal structures and supporting experiments show that editing alters Ca2+ binding. For kinesin-1, a motor protein driving axonal transport, editing regulates transport velocity down microtubules. Seasonal sampling of wild-caught specimens indicates that temperature-dependent editing occurs in the field as well. These data show that A-to-I editing tunes neurophysiological function in response to temperature in octopus and most likely other coleoids.

Squid express conserved ADAR orthologs that possess novel features.

The coleoid cephalopods display unusually extensive mRNA recoding by adenosine deamination, yet the underlying mechanisms are not well understood. Because the adenosine deaminases that act on RNA (ADAR) enzymes catalyze this form of RNA editing, the structure and function of the cephalopod orthologs may provide clues. Recent genome sequencing projects have provided blueprints for the full complement of coleoid cephalopod ADARs. Previous results from our laboratory have shown that squid express an ADAR2 homolog, with two splice variants named sqADAR2a and sqADAR2b and that these messages are extensively edited. Based on octopus and squid genomes, transcriptomes, and cDNA cloning, we discovered that two additional ADAR homologs are expressed in coleoids. The first is orthologous to vertebrate ADAR1. Unlike other ADAR1s, however, it contains a novel N-terminal domain of 641 aa that is predicted to be disordered, contains 67 phosphorylation motifs, and has an amino acid composition that is unusually high in serines and basic amino acids. mRNAs encoding sqADAR1 are themselves extensively edited. A third ADAR-like enzyme, sqADAR/D-like, which is not orthologous to any of the vertebrate isoforms, is also present. Messages encoding sqADAR/D-like are not edited. Studies using recombinant sqADARs suggest that only sqADAR1 and sqADAR2 are active adenosine deaminases, both on perfect duplex dsRNA and on a squid potassium channel mRNA substrate known to be edited in vivo. sqADAR/D-like shows no activity on these substrates. Overall, these results reveal some unique features in sqADARs that may contribute to the high-level RNA recoding observed in cephalopods.

Adaptive Proteome Diversification by Nonsynonymous A-to-I RNA Editing in Coleoid Cephalopods.

RNA editing by the ADAR enzymes converts selected adenosines into inosines, biological mimics for guanosines. By doing so, it alters protein-coding sequences, resulting in novel protein products that diversify the proteome beyond its genomic blueprint. Recoding is exceptionally abundant in the neural tissues of coleoid cephalopods (octopuses, squids, and cuttlefishes), with an over-representation of nonsynonymous edits suggesting positive selection. However, the extent to which proteome diversification by recoding provides an adaptive advantage is not known. It was recently suggested that the role of evolutionarily conserved edits is to compensate for harmful genomic substitutions, and that there is no added value in having an editable codon as compared with a restoration of the preferred genomic allele. Here, we show that this hypothesis fails to explain the evolutionary dynamics of recoding sites in coleoids. Instead, our results indicate that a large fraction of the shared, strongly recoded, sites in coleoids have been selected for proteome diversification, meaning that the fitness of an editable A is higher than an uneditable A or a genomically encoded G.

Profiling the genetic determinants of chromatin accessibility with scalable single-cell CRISPR screens.

CRISPR screens have been used to connect genetic perturbations with changes in gene expression and phenotypes. Here we describe a CRISPR-based, single-cell combinatorial indexing assay for transposase-accessible chromatin (CRISPR-sciATAC) to link genetic perturbations to genome-wide chromatin accessibility in a large number of cells. In human myelogenous leukemia cells, we apply CRISPR-sciATAC to target 105 chromatin-related genes, generating chromatin accessibility data for ~30,000 single cells. We correlate the loss of specific chromatin remodelers with changes in accessibility globally and at the binding sites of individual transcription factors (TFs). For example, we show that loss of the H3K27 methyltransferase EZH2 increases accessibility at heterochromatic regions involved in embryonic development and triggers expression of genes in the HOXA and HOXD clusters. At a subset of regulatory sites, we also analyze changes in nucleosome spacing following the loss of chromatin remodelers. CRISPR-sciATAC is a high-throughput, single-cell method for studying the effect of genetic perturbations on chromatin in normal and disease states.

Spatially regulated editing of genetic information within a neuron.

In eukaryotic cells, with the exception of the specialized genomes of mitochondria and plastids, all genetic information is sequestered within the nucleus. This arrangement imposes constraints on how the information can be tailored for different cellular regions, particularly in cells with complex morphologies like neurons. Although messenger RNAs (mRNAs), and the proteins that they encode, can be differentially sorted between cellular regions, the information itself does not change. RNA editing by adenosine deamination can alter the genome's blueprint by recoding mRNAs; however, this process too is thought to be restricted to the nucleus. In this work, we show that ADAR2 (adenosine deaminase that acts on RNA), an RNA editing enzyme, is expressed outside of the nucleus in squid neurons. Furthermore, purified axoplasm exhibits adenosine-to-inosine activity and can specifically edit adenosines in a known substrate. Finally, a transcriptome-wide analysis of RNA editing reveals that tens of thousands of editing sites (>70% of all sites) are edited more extensively in the squid giant axon than in its cell bodies. These results indicate that within a neuron RNA editing can recode genetic information in a region-specific manner.

Abundant off-target edits from site-directed RNA editing can be reduced by nuclear localization of the editing enzyme.

Site-directed RNA editing (SDRE) is a general strategy for making targeted base changes in RNA molecules. Although the approach is relatively new, several groups, including our own, have been working on its development. The basic strategy has been to couple the catalytic domain of an adenosine (A) to inosine (I) RNA editing enzyme to a guide RNA that is used for targeting. Although highly efficient on-target editing has been reported, off-target events have not been rigorously quantified. In this report we target premature termination codons (PTCs) in messages encoding both a fluorescent reporter protein and the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) protein transiently transfected into human epithelial cells. We demonstrate that while on-target editing is efficient, off-target editing is extensive, both within the targeted message and across the entire transcriptome of the transfected cells. By redirecting the editing enzymes from the cytoplasm to the nucleus, off-target editing is reduced without compromising the on-target editing efficiency. The addition of the E488Q mutation to the editing enzymes, a common strategy for increasing on-target editing efficiency, causes a tremendous increase in off-target editing. These results underscore the need to reduce promiscuity in current approaches to SDRE.

Trade-off between Transcriptome Plasticity and Genome Evolution in Cephalopods.

RNA editing, a post-transcriptional process, allows the diversification of proteomes beyond the genomic blueprint; however it is infrequently used among animals for this purpose. Recent reports suggesting increased levels of RNA editing in squids thus raise the question of the nature and effects of these events. We here show that RNA editing is particularly common in behaviorally sophisticated coleoid cephalopods, with tens of thousands of evolutionarily conserved sites. Editing is enriched in the nervous system, affecting molecules pertinent for excitability and neuronal morphology. The genomic sequence flanking editing sites is highly conserved, suggesting that the process confers a selective advantage. Due to the large number of sites, the surrounding conservation greatly reduces the number of mutations and genomic polymorphisms in protein-coding regions. This trade-off between genome evolution and transcriptome plasticity highlights the importance of RNA recoding as a strategy for diversifying proteins, particularly those associated with neural function. PAPERCLIP.