Modeling Psychiatric Disorder Biology with Stem Cells

Das, Debamitra; Feuer, Kyra L; Wahbeh, Marah H.; Avramopoulos, Dimitrios

doi:10.1007/s11920-020-01148-1

Cited by 29 publications

(39 citation statements)

References 329 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To demonstrate this function, we utilized DECODE to predict compact enhancers in NPCs, which play important roles in neuro-development and have been implicated in a wide variety of psychiatric disorders (Castrén, 2014; Das et al ., 2020). Specifically, we utilized the five epigenetic marks on NPCs to predict enhancers using our DECODE framework trained on all available data (see details in Methods 2.4).…”

Section: Resultsmentioning

confidence: 99%

DECODE: ADeep-learning Framework forCondensing Enhancers and Refining Boundaries with Large-scale Functional Assays

Chen

Zhang

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

SummaryMapping distal regulatory elements, such as enhancers, is the cornerstone for investigating genome evolution, understanding critical biological functions, and ultimately elucidating how genetic variations may influence diseases. Previous enhancer prediction methods have used either unsupervised approaches or supervised methods with limited training data. Moreover, past approaches have operationalized enhancer discovery as a binary classification problem without accurate enhancer boundary detection, producing low-resolution annotations with redundant regions and reducing the statistical power for downstream analyses (e.g., causal variant mapping and functional validations). Here, we addressed these challenges via a two-step model called DECODE. First, we employed direct enhancer activity readouts from novel functional characterization assays, such as STARR-seq, to train a deep neural network classifier for accurate cell-type-specific enhancer prediction. Second, to improve the annotation resolution (∼500 bp), we implemented a weakly-supervised object detection framework for enhancer localization with precise boundary detection (at 10 bp resolution) using gradient-weighted class activation mapping.ResultsOur DECODE binary classifier outperformed the state-of-the-art enhancer prediction methods by 24% in transgenic mouse validation. Further, DECODE object detection can condense enhancer annotations to only 12.6% of the original size, while still reporting higher conservation scores and genome-wide association study variant enrichments. Overall, DECODE improves the efficiency of regulatory element mapping with graphic processing units for deep-learning applications and is a powerful tool for enhancer prediction and boundary localization.Contactpi@gersteinlab.org

show abstract

Section: Resultsmentioning

confidence: 99%

DECODE: ADeep-learning Framework forCondensing Enhancers and Refining Boundaries with Large-scale Functional Assays

Chen

Zhang

Liu

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Our results highlight the complexity of the interplay between population genetics and regulatory variation, which creates unpredictable relationships between the effects of variants of gene expression and that of gene expression on the risk. As it becomes increasingly common to manipulate the genome in targeted ways in order to understand the biology basis of disease risk represented by GWAS variants [Das and others 2020], understanding this interplay is increasingly important. At the same time however, this opens the possibility that small observed effects of variants on risk might underestimate the effects of individual gene expression levels on the risk, which in turn could open new possibilities for therapeutic interventions.…”

Section: Discussionmentioning

confidence: 99%

“…The consequence of this is that if a disease risk allele increases the expression of the gene it does not necessarily follow that increased expression translates to higher risk. This is of great importance as disease-modeling studies often perturb gene expression to study the outcome and understand the biology of disease [Das and others 2020; Hill and others 2012; Schrode and others 2019; Yang and others 2018]. It is therefore necessary to seek formal evidence that a specific gene’s expression mediates disease risk and in which direction, if we want to have an accurate list of the genes and a correct understanding of their role in disease.…”

Section: Introductionmentioning

confidence: 99%

Schizophrenia Risk Alleles Often Affect the Expression of Many Genes and Each Gene May Have a Different Effect on The Risk; A Mediation Analysis

Peng

Bader

Avramopoulos

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Variants identified by Genome wide association studies (GWAS) are often expression quantitative trait loci (eQTLs), suggesting they are proxies or are themselves regulatory.Additionally eQTL analyses show that variants often affect more than one gene. Lacking data on many tissue types, developmental time points and homogeneous cell types, the extent of this one-to-many relationship is underestimated. This raises questions on whether a disease eQTL target gene explains the genetic association or is a by-stander.It also puts into question the direction of the effect of a gene's expression on the risk, since the many genes regulated by the same variant may have opposing effects, imperfectly balancing each other. We used two brain gene expression datasets (CommonMind and BrainSeq) for a mediation analysis of schizophrenia-associated variants. We find that indeed eQTL targets often mediate risk but the direction in which expression affects risk is often different from the direction in which the risk allele changes expression. Of 38 genes significant for mediation in both datasets 33 showed consistent direction (Chi 2 test P=6*10 -6 ) and for 15 of them (45%) the expression change associated with the risk allele was protective, which suggests the likely presence of other target genes with overriding effects. Our results identify specific risk mediating genes and suggest caution in interpreting the biological consequences of targeted modifications of gene expression, as not all eQTL targets may be relevant to disease and those that are might have different than expected directions.

show abstract

“…Several methods have also been developed using the standard CRISPR/Cas9 editing tools to facilitate scarless editing 20 . However, these methods have one or more of the following limitations: In recent years, alternative Cas9-based editors were developed that allow for scarless editing without producing DSBs, such as base editors 21; 22 and prime editors 23 .…”

Section: Limitations Of Current Scarless Editing Methodsmentioning

confidence: 99%

“…Several methods have also been developed using the standard CRISPR/Cas9 editing tools to facilitate scarless editing 20 . However, these methods have one or more of the following limitations: ( i ) low editing efficiency (1-5%); ( ii ) laborious and/or expensive clone screening; ( iii ) dependence on a PAM site near the target base; ( iv ) requirement of a specific genotype in the starting cell line (when editing the PAM site); ( v ) inconsistent experimental procedures for edited and unedited clones in experiments comparing isogenic clones.…”

Section: Introductionmentioning

confidence: 99%

CRISPR Del/Rei: A simple, flexible, and efficient pipeline for scarless genome editing

Wahbeh

Feuer

Abdollahi³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Scarless genome editing is an important tool for the accurate recapitulation of genetic variation in human disease models. Various CRISPR/Cas9-based scarless editing methods have been reported. However, some of these methods have low editing efficiency (1-5%) and require manual selection of hundreds of clones to reach the desired number. Other protocols use large selection cassettes with laborious vector assembly and specialized reagents and equipment, or have poorly understood off-target effects. To address these limitations, we developed a simple, highly efficient scarless editing strategy to edit DNA sequences in induced pluripotent stem cells, which we call CRISPR Del/Rei. This novel editing strategy consists of a two-step deletion-reinsertion strategy that produces isogenic clones in ~8 weeks using accessible, user-friendly reagents. The editing efficiency ranges from ~15–100% for Step 1 and ~5–20% for Step 2 after selection, which greatly reduces the amount of required manual clone isolation. Screening the transfected bulk cells and the individual clones is rapid and simple, consisting of PCR and gel electrophoresis. Despite the two editing steps, off-target effects are rare. Additionally, the experiment is well-controlled because the same protocol generates isogenic clones carrying all variant alleles. In this way, CRISPR Del/Rei serves as a valuable addition to the evolving CRISPR/Cas9 gene-editing toolset.

show abstract

Modeling Psychiatric Disorder Biology with Stem Cells

Cited by 29 publications

References 329 publications

DECODE: ADeep-learning Framework forCondensing Enhancers and Refining Boundaries with Large-scale Functional Assays

DECODE: ADeep-learning Framework forCondensing Enhancers and Refining Boundaries with Large-scale Functional Assays

Schizophrenia Risk Alleles Often Affect the Expression of Many Genes and Each Gene May Have a Different Effect on The Risk; A Mediation Analysis

CRISPR Del/Rei: A simple, flexible, and efficient pipeline for scarless genome editing

Contact Info

Product

Resources

About