Biology subjects

Zimmermann, M. T.

Publications and source records attributed to Zimmermann, M. T..

15 recordsLinked to original sources

bioRxiv2026

Defining Pseudo-Haplotype Analysis Reveals Multi-Gene Genetic Pattern Across BAF Chromatin Remodeling Complexes

BRG1-associated factor (BAF) is a crucial chromatin remodeling complex. Variants in genes encoding BAF complex components cause human diseases, including cancers and developmental disorders. However, the genetic diversity and variant co-occurrence patterns within BAF genes remain incompletely understood. It is feasible, though largely untested, that rare patterns of common variations could alter function similarly to rare deleterious variants. Further, there is no modern census of how often individual people simultaneously carry multiple rare and common variations, nor means for genomics practitioners to assess their combined effects. Approaches are needed to characterize complete sequences from individual samples. In this study, we introduce a pseudo-haplotype analysis (PHA) framework, combining multiple protein-coding sequence variants, observed concurrently within individual samples, into discrete BAF patterns. In this cohort, 78.44% of pseudo-haplotype (PH) copies carry at least one BAF coding variation. Among these, 56.18% contain at least two distinct variants, and 32.39% contain three or more, indicating a substantial burden of multi-variant configurations across individuals. Notably, 25.30% of unique PHs are observed only once, highlighting a considerable proportion of people who are affected by rare or private combinations of genetic variations. We identify multiple significant (FDR < 0.05) co-occurrence combinations across global populations. These findings underscore the importance of considering population-specific genetic structures, and complete individual variant configurations when investigating disease associations and genetic mechanisms. Our approach provides a generalizable framework for characterizing multi-variant architectures within chromatin remodeling genes at a population scale, with potential applications in elucidating disease etiology and advancing precision medicine.

bioinformatics↗

bioRxiv2026

Defining the molecular tolerance-to-damage landscape of SMARCA4 helicase genetic alterations

Evaluating the impact of genomic variation is essential for identifying underlying mechanistic causes of human diseases. The spectrum of neurodevelopmental disorders is driven by diverse genetic alterations with genes like SMARCA4 being prototypical examples. There have been significant hurdles to implementing the protein-specific and mechanism-informed variation effect predictors that are anticipated to have the highest yield of mechanistic information. Yet, there is a pressing need, for example, within SMARCA4 where 98% of the 2780 reported variants lack a disposition and remain of uncertain significance (VUS). Further, the field has yet to identify each variants specific molecular mechanism, which will inform targeted therapeutic development strategies. In this study we developed a mechanistic structure-informed helicase-specific variant effect predictor by leveraging diverse information with state-specific calculations. Our approach has 100% recall of pathogenic variants while classifying 87.23% of VUS into damaging (55.74%, n=262) versus tolerated effects (31.49%, n=148), including those with conflicting interpretations. This analysis reveals significant enrichment of integrated functional metrics, such as conservation and solvent exposure, that parallel allele frequences in health populations, and emphasizes the robustness of the method. Thus, we have demonstrated a novel approach for the development of mechanism-informed protein-specific interpretation of human genetic information.

genetics↗

bioRxiv2026

ARID1B damaging variants from more than one million genomes, cause human diseases by impairing protein-protein interactions, stability, and regulation

ARID1B is the most frequently de novo altered gene across a spectrum of human neurodevelopmental disorders and cancers. We found 1,456 missense variants in ARID1Bs C-terminal domains, 94% of which are clinically uninterpreted and 59% of which are observed in human disorders and cancers. Using integrative modeling of ARID1B cBAF and DNA bound structures, we calculate how these variants impact key functional sites, structural stability, and topological features, finding that 617 variants clearly impair ARID1Bs interactions within BAF, interactions with DNA, fold stability, or post-translational regulation. Our study enables scalable and precise variant interpretation by illuminating the molecular mechanisms by which missense variants damage ARID1B function, with commonalities across malignancies and congenital anomalies mutually informing therapeutic developments.

genetics↗

bioRxiv2025

Tissue-Specific Co-Expression Patterns of BAF Complexes Provide Regulatory Insights Across Human Tissues with Implications for Endocrine and Non-Endocrine Functions

BRG1/BRM-associated factor (BAF) chromatin remodeling complexes are essential for normal endocrine function and are implicated in various metabolic and developmental disorders. However, the full range of chromatin-based regulatory modules in endocrine development remains unclear. We developed a computational pipeline to analyze bulk RNA-seq data from 54 human tissues and constructed tissue-specific co-expression networks for 30 core BAF complex genes. Weighted gene co-expression network analysis (WGCNA) and Louvain clustering identified gene modules for each tissue, which we compared to 46 curated BAF subcomplex gene sets using Jaccard similarity. In metabolically active non-endocrine tissues (kidney, skeletal muscle, vasculature, fibroblasts), we observed strong co-expression with canonical BAF (cBAF) and polybromo-associated BAF (pBAF) modules. Central nervous system tissues were dominated by neuron-specific BAF (nBAF) modules. Endocrine tissues (e.g., thyroid, adrenal) and gastrointestinal epithelia displayed co-expression profiles resembling smooth muscle-like BAF and pBAF modules, suggesting chromatin programs that integrate hormone secretion with contractile and barrier functions. These patterns show that each tissue exhibits a distinct, non-random combination of BAF subcomplexes, potentially reflecting its functional chromatin state. Our results demonstrate that tissue-specific gene expression profiling can reveal differences in protein complex regulation. The modular deployment of BAF chromatin remodeling complexes appears tailored to the functional demands of each organ. This study lays a foundation for further investigation of epigenetic regulation in endocrine development and disease and provides a framework for identifying tissue-specific chromatin remodeling strategies. Plain Language Summary (optional)The US National Institutes of Health have invested in large-scale measurements of how different human body tissues use their genetic material. This study pilots use of these data to understand better how various genes come together to form nanomachines that in turn regulate the same genetic material.

bioinformatics↗

bioRxiv2025

A Multi-State Structural Genomics Approach Enables Large-Scale, Mechanistic, and Context-Specific Classification of ABCC6 Genetic Variants Implicated in Calcification Diseases

PurposeGenetic variation in ATP Binding Cassette Subfamily C Member 6 (ABCC6) can cause both pseudoxanthoma elasticum (PXE) and generalized arterial calcification of infancy (GACI). Despite both diseases being rare, there are already 930 distinct missense variants in ABCC6 reported, 87% of which are of uncertain clinical significance (VUS). New approaches are needed to interpret and classify these VUS mechanistically. MethodsWe developed 3D protein models of ABCC6 in three functionally relevant conformations to calculate the structural effects of variants and identify 3D mutational hotspots. With this and additional functional information, we categorized variants in a mechanistic ontology based on which critical functions of ABCC6 they impact. We then compared PXE and GACI -associated variants. ResultsWe identified two three-dimensional hotspots of pathogenic variants and six specific functions of ABCC6 which variants impact. From this, we propose a mechanism for pathogenicity for 41% of VUS according to their impacted function, 30 of which could be reclassified as Likely Pathogenic from our non-clinical data. Finally, we found slight differences between PXE and GACI-associated variants. ConclusionThe mechanistic information we present will guide future research to better address calcification disorders and understand genetic variants. Further, our VUS reclassification will improve the diagnosis of ABCC6-driven diseases, shortening diagnostic odysseys. We believe that computational structural genomics approaches will soon take prominence in genomics data interpretation. HighlightsVariant 3D hotspot detection and effect clustering, GACI and PXE Variant Comparison, consistent ability to (re)categorize known pathogenic variants, and population genetics differences.

genetics↗

bioRxiv2025

An Integrative Multitiered Computational Analysis for Better Understanding the Structure and Function of 85 Miniproteins

BackgroundMiniproteins, defined as polypeptides containing fewer than 50 amino acids, have recently elicited significant interest due to an emerging understanding of their diverse roles in fundamental biological processes. In addition, miniprotein dysregulation underlies human diseases and is a significant focus for biotechnology and drug development. Notably, the human genome project revealed the existence of many novel miniproteins, most of which remain uncharacterized. This study reports an approach for analyzing and scoring previously uncharacterized miniproteins by integrating knowledge from classic sequence-based bioinformatics, computational biophysics, and system biology annotations. Our results demonstrate that these approaches provide novel information on the structure-function relationship of these molecules with a particular focus on their biomedical relevance. MethodsWe identified 85 human miniproteins using a simple multi-tier approach. First, we performed a sequence-based analysis of these proteins using several algorithms to identify regions of structural and functional importance. Protein-protein interactions and gene ontology annotations were used to analyze miniprotein function. Then, we predicted miniprotein three-dimensional structures using AI-based methods and peptide modeling to determine their relative yields for these understudied polymers. Subsequently, we used several computational biophysics methods and structure-based calculations to annotate and evaluate results from both algorithms. ResultsWe find several relations between predicted structure and functional properties to assign these proteins to several groups with similar properties. Sequence-based analysis leads us to identify motifs and residues that link structure-to-function for most of these proteins. We suggest novel miniprotein functions, such as thymosin beta proteins regulating the shelterin complex through TERF1 and POT1 interactions, FAM86JP and FAM66E participating in endocytic processes, and BAGE1 influencing chromatin remodeling through interaction with nuclear proteins. Further, known functions of miniproteins, such as STRIT1, STMP1, and SLN, were supported. Finally, structure-based scoring led us to build 3D models that provided complementary information to ontologies. We identify that structural propensity is not strictly dependent on polymer length. In fact, in this dataset, peptide-based algorithms may have advantages over AI-based algorithms for certain groups of miniproteins. ConclusionThis analytic approach and resulting identification and annotation of miniproteins adds much to what is currently known about miniproteins. Our determination of novel properties of miniproteins bears significant mechanistic and biomedical relevance. We propose novel functions of miniproteins, which expands our understanding of their potential roles in cellular processes. And, we practically identify which sequence and structure-based tools provide the most information, aiding future studies of miniproteins.

bioinformatics↗

bioRxiv2024

Advancing Variant Phenotyping in Myelodysplastic Syndromes via Computational Genomics of Mitochondrial Enzyme Complexes

Mitochondria are essential cellular organelles that play critical roles in hematological disorders. Recurrent mutations in mitochondrial DNA (mtDNA) have been identified in patients with myelodysplastic syndromes (MDS) and serve as significant prognostic indicators for their outcomes following allogeneic hematopoietic stem-cell transplantation (allo-HCT). However, the biological mechanisms of mtDNA mutations remain unclear. The current study utilizes computational structural genomics to improve our understanding of pathogenic variants in mitochondria-encoded genes. This emerging genomics discipline employs structural models, molecular mechanic calculations, and accelerated molecular dynamic simulations to analyze gene products, focusing on their structures and motions that determine their function. We applied this methodology to perform deep variant phenotyping of entire mitochondria-encoded protein complexes associated with the pathobiology of MDS and their prognosis after HCT. Our results demonstrate that this approach significantly outperforms conventional analytical methods, providing enhanced and more accurate information to support the potential pathogenicity of these variants and better infer their dysfunctional mechanisms. We conclude that the adoption and further expansion of computational structural genomics approaches, as applied to the mitochondrial genome, have the potential to significantly increase our understanding of molecular mechanisms underlying the disease. Our study lays a foundation for translating mitochondrial biology into clinical applications, which is of significant mechanistic and biomedical relevance and should be considered in modern biomedical research.

genomics↗

bioRxiv2024

EHMT2 Inactivation in Pancreatic Epithelial Cells Shapes the Transcriptional Landscape and Inflammation Response of the Whole Pancreas

The Euchromatic Histone Methyl Transferase Protein 2 (EHMT2), also known as G9a, deposits transcriptionally repressive chromatin marks that play pivotal roles in the maturation and homeostasis of multiple organs. Recently, we have shown that EHMT2 inactivation alters growth and immune gene expression networks, antagonizing KRAS-mediated pancreatic cancer initiation and promotion. Here, we elucidate the essential role of EHMT2 in maintaining a transcriptional landscape that protects organs from inflammation. Comparative RNA-seq studies between normal postnatal and young adult pancreatic tissue from EHMT2 conditional knockout animals (EHMT2fl/fl) targeted to the exocrine pancreatic epithelial cells (Pdx1-Cre and P48Cre/+), reveal alterations in gene expression networks in the whole organ related to injury-inflammation-repair, suggesting an increased predisposition to damage. Thus, we induced an inflammation repair response in the EHMT2fl/flpancreas and used a data science-based approach to integrate RNA-seq-derived pathways and networks, deconvolution digital cytology, and spatial transcriptomics. We also analyzed the tissue response to damage at the morphological, biochemical, and molecular pathology levels. The EHMT2fl/fl pancreas displays an enhanced injury-inflammation-repair response, offering insights into fundamental molecular and cellular mechanisms involved in this process. More importantly, these data show that conditional EHMT2 inactivation in exocrine cells reprograms the local environment to recruit mesenchymal and immunological cells needed to mount an increased inflammatory response. Mechanistically, this response is an enhanced injury-inflammation-repair reaction with a small contribution of specific EHMT2-regulated transcripts. Thus, this new knowledge extends the mechanisms underlying the role of the EHMT2-mediated pathway in suppressing pancreatic cancer initiation and modulating inflammatory pancreatic diseases.

cancer biology↗

bioRxiv2023

Assessing Protein Surface-Based Scoring for Interpreting Genomic Variants

Clinical genomics sequencing is rapidly expanding the number of variants that need to be functionally elucidated. Interpreting genetic variants (i.e., mutations) usually begins by identifying how they affect protein-coding sequences. Still, the 3D protein molecule is rarely considered for large-scale variant analysis, nor for their effects on how proteins interact with each other and their environment. We propose a standardized approach to scoring protein surface property changes as a new dimension for functionally and mechanistically interpreting genomic variants. Further, it directs hypothesis generation for functional genomics research to learn more about the encoded proteins function. We developed a novel method leveraging 3D structures and time-dependent simulations to score and statistically evaluate protein surface property changes. We evaluated positive controls composed of 8 thermophilic versus mesophilic orthologs and variants that experimentally change protein solubility, which all showed large and statistically significant differences in charge distribution (p<0.01). We scored static 3D structures and dynamic ensembles for 43 independent variants (23 pathogenic and 20 uninterpreted) across four proteins. Focusing on the potassium ion channel, KCNK9, average local surface potential shifts were 0.41 kBT/ec with an average p-value of 1x10-2. In contrast, dynamic ensemble shifts averaged 1.15 kBT/ec and an average p-value of 1x10-5, enabling the identification of changes far from mutated sites. This study demonstrates that an objective assessment of how mutations affect electrostatic distributions of protein surfaces can aid in interpreting genomic variants discovered through clinical genomics sequencing.

bioinformatics↗

bioRxiv2023

A Multi-Layered Computational Structural Genomics Approach Enhances Domain-Specific Interpretation of Kleefstra Syndrome Variants in EHMT1

This study investigates the functional significance of assorted variants of uncertain significance (VUS) in euchromatic histone lysine methyltransferase 1 (EHMT1), which is critical for early development and normal physiology. EHMT1 mutations cause Kleefstra syndrome and are linked to various human cancers. However, accurate functional interpretation of these variants are yet to be made, limiting diagnoses and future research. To overcome this, we integrate conventional tools for variant calling with computational biophysics and biochemistry to conduct multi-layered mechanistic analyses of the SET catalytic domain of EHMT1, which is critical for this protein function. We use molecular mechanics and molecular dynamics (MD)-based metrics to analyze the SET domain structure and functional motions resulting from 97 Kleefstra syndrome missense variants within this domain. Our approach allows us to classify the variants in a mechanistic manner into SV (Structural Variant), DV (Dynamic Variant), SDV (Structural and Dynamic Variant), and VUS (Variant of Uncertain Significance). Our findings reveal that the damaging variants are mostly mapped around the active site, substrate binding site, and pre-SET regions. Overall, we report an improvement for this method over conventional tools for variant interpretation and simultaneously provide a molecular mechanism of variant dysfunction.

genomics↗

bioRxiv2023

Beyond Structural Bioinformatics for Genomics with Dynamics Characterization of an Expanded KRAS Mutational Landscape

ABSTRACTCurrent capabilities in genomic sequencing outpace functional interpretations. Our previous work showed that 3D protein structure calculations enhance mechanistic understanding of genetic variation in sequenced tumors and patients with rare diseases. The KRAS GTPase is among the critical genetic factors driving cancer and germline conditions. Because KRAS-altered tumors frequently harbor one of three classic hotspot mutations, nearly all studies have focused on these mutations, leaving significant functional ambiguity across the broader KRAS genomic landscape observed in cancer and non-cancer diseases. Herein, we extend structural bioinformatics with molecular simulations to study an expanded landscape of 86 KRAS mutations. We identify multiple coordinated changes strongly associated with experimentally established KRAS biophysical and biochemical properties. The patterns we observe span hotspot and non-hotspot alterations, which can all dysregulate Switch regions, producing mutation-restricted conformations with different effector binding propensities. We experimentally measured mutation thermostability and identified shared and distinct patterns with simulations. Our results indicate mutation-specific conformations which show potential for future research into how these alterations reverberate into different molecular and cellular functions. The data we present is not predictable using current genomic tools, demonstrating the added functional information derived from molecular simulations for interpreting human genetic variation. Key PointsPlease provide 3 bullet points summarizing the manuscripts contribution to the field (100 characters max per point) 1) We functionally grouped 86 distinct KRAS mutations using MD scores, demonstrating scalability for genomics 2) MD-based groups explain experimental differences and mechanistic information about mutant proteins 3) Demonstrated added functional information from simulations for interpreting human genetic variation

genomics↗

bioRxiv2022

Microglia influence neurofilament deposition in ALS iPSC-derived motor neurons

Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease in which upper and lower motor neuron loss is the primary phenotype, leading to muscle weakness and wasting, respiratory failure, and death. Although a portion of ALS cases are linked to one of over 50 unique genes, the vast majority of cases are sporadic in nature. However, the mechanisms underlying the motor neuron loss in either familial or sporadic ALS are not entirely clear. Here we used induced pluripotent stem cells derived from a set of identical twin brothers discordant for ALS to assess the role of astrocytes and microglia on the expression and accumulation of neurofilament proteins in motor neurons. We found that motor neurons derived from the affected twin exhibited increased transcript levels of all three neurofilament isoforms and increased expression of phosphorylated neurofilament puncta. We further found that treatment of the motor neurons with astrocyte conditioned medium and microglial conditioned medium significantly impacted neurofilament deposition. Together, these data suggest that glial-secreted factors can alter neurofilament pathology in ALS iPSC-derived motor neurons.

neuroscience↗

bioRxiv2020

Polymorphisms in STING affect human innate immune responses to poxviruses

We conducted a large genome-wide association study (GWAS) of the immune responses to primary smallpox vaccination in a combined cohort of > 1,600 subjects. We identified a cluster of SNPs on chromosome 5 (5q31.2) that were significantly associated (p-value: 1.3 x 10-12 - 1.5x10-36) with IFN response to in vitro poxvirus stimulation. Examination of these SNPs led to the functional testing of rs1131769, a non-synonymous SNP in TMEM173 causing an Arg-to-His change at position 232 in the STING protein--a major regulator of innate immune responses to viral infections. Our findings demonstrate important functional differences between the two alleles, where the major allele (R232) more effectively induces IFN secretion. Molecular modeling of both alleles identified altered ligand binding characteristics between the two variants, providing a potential mechanism underlying differences in inter-individual responses to poxvirus vaccination. Our data demonstrate that possession of the H232 variant impairs STING-mediated innate immunity to poxviruses. These results clarify prior studies evaluating functional effects of genetic variants in TMEM173 and provide novel data regarding genetic control of poxvirus immunity. Contribution to the FieldHere we report that a single nucleotide non-synonymous polymorphism in the TMEM173 gene encodes for a STING variant conferring a reduced IFN stimulated response compared to wild type. Our results suggest that, upon binding of the STING H232 variant to its ligand, activation of downstream signaling proteins is impaired, resulting in decreased production of IFN and a weaker interferon-stimulated gene response. Molecular modeling indicates that the diminished functional activity of this variant is likely due to an altered physical structure of the STING protein. STING controls the innate, type I IFN response to double-stranded DNA and cyclic dinucleotides. Individuals with the H232 variant of STING have a much weaker innate immune response to vaccinia virus. Our data help resolve ongoing controversies regarding the role of genetic variants in STING function. Because STING plays an important role in our immune response to DNA viruses and bacteria, our results can be used to predict who will and will not respond to vaccines and treatments, and to design more effective vaccine candidates. Given the role of the STING protein in innate responses to DNA viruses and bacterial pathogens, these data may also be useful in developing novel treatment options for multiple infectious diseases.

immunology↗

bioRxiv2019

Integration of Multi-level Molecular Scoring for the Interpretation of RAS-Family Genetic Variation

Protein-coding genetic variants are the first considered in most studies and Precision Medicine workflows, but their interpretation is primarily driven by DNA sequence-based analytical tools and annotations. Thus, more specific and mechanistic interpretations should be attainable by integrating DNA-based scores with scores from the protein 3D structure. However, reliable and reproducible standardization of methods that use 3D structure for genomic variation is still lacking. Further, we believe that the current paradigm of aiming to directly predict the pathogenicity of variants skips the critical step of inferring, with precision, molecular mechanisms of dysfunction. Thus, we report herein the development and evaluation of single and composite 3D structure-based scores and their integration with protein and DNA sequence-based scores to better understand not only if a genomic variant alters a protein, but how. We believe this is a critical step for understanding mechanistic changes due to genomic variants, designing functional validation tests, and for improving disease classifications. We applied this approach to the RAS gene family encoding seven distinct proteins and their 935 unique missense variants present somatically in cancer, in rare diseases (termed RASopathies), and in the currently healthy adult population. This knowledge shows that protein structure-based scores are distinct from information available from genomic annotation, that they are useful for interpreting genomic variants, and they should be taken into consideration in future guidelines for genomic data interpretation.\n\nSignificance StatementGenetic information from patients is a powerful data type for understanding individual differences in disease risk and treatment, but most of the genetic variation we observe has no mechanistic interpretation. This lack of interpretation limits the use of genomics data in clinical care. Standard methods for genomics data interpretation take advantage of annotations available for the human reference genome, but they do not consider the 3D protein molecule. We believe that changes to the 3D molecule must be considered, to augment current practice and lead to more precise interpretation. In this work, we present our initial process for systematic multi-level molecular scores, including 3D, to interrogate 935 RAS-family variants that are relevant in both cancer and rare diseases.

genetics↗

bioRxiv2019

Discovery, Expression, Cellular Localization, and Molecular Properties of a Novel, Alternative Spliced HP1g Isoform, Lacking the Chromoshadow Domain

By reading the H3K9Me3 mark through their N-terminal chromodomain (CD), HP1 proteins play a significant role in cancer-associated processes, including cell proliferation, differentiation, chromosomal stability, and DNA repair. Here, we used a combination of bioinformatics-based methodologies, as well as experimentally-derived datasets, that reveal the existence of a novel short HP1{gamma} (CBX3) isoform, named here sHP1{gamma}, generated by alternative splicing of the CBX3 locus. The sHP1{gamma} mRNA encodes a protein composed of 101 residues and lacks the C-terminal chromoshadow domain (CSD) that is required for dimerization and heterodimerization in the previously described 183 a. a HP1{gamma} protein. Fold recognition, order-to-disorder calculations, threading, homology-based molecular modeling, docking, and molecular dynamic simulations show that the sHP1{gamma} is comprised of a CD flanked by intrinsically disordered regions (IDRs) with an IDR-CD-IDR domain organization and likely retains the ability to bind to the H3K9Me3. Both qPCR analyses and mRNA-seq data derived from large-scale studies confirmed that sHP1{gamma} mRNA is expressed in the majority of human tissues at approximately constant ratios with the chromoshadow domain containing isoform. However, sHP1{gamma} mRNA levels appear to be dysregulated in different cancer types. Thus, our data supports the notion that, due to the existence of functionally different isoforms, the regulation of HP1{gamma}-mediated functions is more complex than previously anticipated.

molecular biology↗