Abstract 3134: Calling somatic variants from UG100 data using deep learning

UG100 is a novel next-generation sequencing platform that combines high throughput with significantly lower sequencing cost. Previous studies have demonstrated broad applicability of UG100 data for whole-genome germline variant calling, single cell transcriptomics and whole-genome methylation analysis, as well as for recalling cancer signatures from cfDNA at very low fraction of circulating tumor DNA.

Somatic variant calling is a natural application for this platform as it can benefit from lower sequencing cost to enable deeper sequencing coverage. Here, we describe the implementation and evaluation of a somatic calling pipeline from UG100 whole genome sequence data. Since deep-learning-based variant calling methods currently outperform statistical variant calling methods for germline variant calling on UG100 data, we cast somatic variant calling as a classification problem.

Specifically, we trained a classifier to distinguish if a candidate at a particular location is a somatic variant or a sequencing error. We used a version of DeepVariant optimized for UG100 data to train the deep-learning classifier in three scenarios: tumor only, tumor with an unmatched background sample and matched tumor-normal samples. The labeled truth set for training was generated by mixing whole genome sequenced samples from the genome-in-a-bottle project in a wide range of proportions (0-100% mixing ratio) to simulate various allele frequencies, with an average genome coverage of 100x. The tumor/normal model was the best-performing of the three models with a recall of >98% for SNPs and 90% for Indels at allele fraction > 10%. Notably, the model also showed high specificity as well with 16 false positive SNPs and 19 false positive indels at AF over 10% called on the chromosome that was not part of the training (chr20).

We then applied the model for calling from the WGS data on three well characterized pairs of matched tumor and normal cell lines: HCC1143, COLO829 and HCC1395. We evaluated the performance on the pre-defined UG-HCR (Ultima Genomics - High Confidence Region), which includes 95% of the human genome. DeepVariant models performed very well on calling SNPs (>92% recall at allele frequencies above 10%) and indels (>90% recall). The calls were also highly specific, with less than 1/Mb variants absent in the ground truth across the UG-HCR.

Lastly, we applied the models to 8 unpaired cell lines with known driver mutations and observed that we call 34/34 driver mutations of length <=20 bp that appear in COSMIC (100% recall).

We expect the UG100 sequencer to become an important tool for somatic genome analysis and to enable deep whole-genome sequencing to become a routine assay in clinical oncology.

Citation Format: Maya Levy, Doron Shem-Tov, Hila Benjamin, Sima Benjamin, Ilya Soifer, Shlomit Gilad, Danit Lebanony, Nika Iremadze, Eti Meiri, Doron Lipson, Omer Barad. Calling somatic variants from UG100 data using deep learning [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3134.

Novel insights from a multiomics dissection of the hayflick limit

The process wherein dividing cells exhaust proliferative capacity and enter into replicative senescence has become a prominent model for cellular aging in vitro. Despite decades of study, this cellular state is not fully understood in culture and even much less so during aging. Here, we revisit Leonard Hayflick’s original observation of replicative senescence in WI-38 human lung fibroblasts equipped with a battery of modern techniques including RNA-seq, single-cell RNA-seq, proteomics, metabolomics, and ATAC-seq. We find evidence that the transition to a senescent state manifests early, increases gradually, and corresponds to a concomitant global increase in DNA accessibility in nucleolar and lamin associated domains. Furthermore, we demonstrate that senescent WI-38 cells acquire a striking resemblance to myofibroblasts in a process similar to the epithelial to mesenchymal transition (EMT) that is regulated by t YAP1/TEAD1 and TGF-β2. Lastly, we show that verteporfin inhibition of YAP1/TEAD1 activity in aged WI-38 cells robustly attenuates this gene expression program.

Loss of major nutrient sensing and signaling pathways suppresses starvation lethality in electron transport chain mutants

The electron transport chain (ETC) is a well-studied and highly conserved metabolic pathway that produces ATP through generation of a proton gradient across the inner mitochondrial membrane coupled to oxidative phosphorylation. ETC mutations are associated with a wide array of human disease conditions and to aging-related phenotypes in a number of different organisms. In this study, we sought to better understand the role of the ETC in aging using a yeast model. A panel of ETC mutant strains that fail to survive starvation was used to isolate suppressor mutants that survive. These suppressors tend to fall into major nutrient sensing and signaling pathways, suggesting that the ETC is involved in proper starvation signaling to these pathways in yeast. These suppressors also partially restore ETC-associated gene expression and pH homeostasis defects, though it remains unclear whether these phenotypes directly cause the suppression or are simply effects. This work further highlights the complex cellular network connections between metabolic pathways and signaling events in the cell and their potential roles in aging and age-related diseases.

Modeling Multiplexed Images with Spatial-LDA Reveals Novel Tissue Microenvironments

Recent in situ multiplexed profiling techniques provide insight into microenvironment formation, maintenance, and transformation through a lens of localized cellular phenotype distribution. In this article, we introduce a method for recovering signatures of microenvironments from such data. We use topic models to identify characteristic cell types overrepresented in neighborhoods that serve as proxies for microenvironment. Furthermore, by assuming spatial coherence among neighboring microenvironments our model limits the number of parameters that need to be learned and permits data-driven decisions about the size of cellular neighborhoods. We apply this method to uncover anatomically known structures in mouse spleen—identifying distinct population of spleen B cells that are defined by their characteristic neighborhoods. Next, we apply the method to a dataset of triple-negative breast cancer tumors from 41 patients to study the structure of tumor-immune boundary. We uncover previously reported tumor-immune microenvironment near the tumor-immune boundary enriched for immune cells with high Indoleamine-pyrrole 2,3-dioxygenase (IDO) and Programmed death-ligand 1 (PD-L1) and a novel, immunosuppressed, microenvironment-enriched for cells expressing CD45 and FoxP3.

A Visual Framework for Classifying Determinants of Cell Size

Cells control their size by coordinating cell cycle progression with volume growth. Size control is typically studied at specific cell cycle transitions that are delayed or accelerated depending on size. This focus is well suited for revealing mechanisms acting at these transitions, but neglects the dynamics in other cell cycle phases, and is therefore inherently limited for studying how the characteristic cell size is determined. We address this limitation through a formalism that intuitively visualizes the characteristic size emerging from integrated cell cycle dynamics of individual cells. Applying this formalism to budding yeast, we describe the contributions of the un-budded (G1) and budded (S-G2-M) phase to size adjustments following environmental or genetic perturbations. We show that although the budded phase can be perturbed with little consequences for G1 dynamics, perturbations in G1 propagate to the budded phase. Our study provides an integrated view on cell size determinants in budding yeast.

•

An intuitive visualization framework for cell size control is described

•

Cell size control in different environments or mutant backgrounds can be compared

•

Mutual dependencies between size control at different cell cycle phases are described

MANF regulates metabolic and immune homeostasis in ageing and protects against liver damage

Aging is accompanied by altered intercellular communication, deregulated metabolic function, and inflammation. Interventions that restore a youthful state delay or reverse these processes, prompting the search for systemic regulators of metabolic and immune homeostasis. Here we identify MANF, a secreted stress-response protein with immune modulatory properties, as an evolutionarily conserved regulator of systemic and in particular liver metabolic homeostasis. We show that MANF levels decline with age in flies, mice and humans, and MANF overexpression extends lifespan in flies. MANF deficient flies exhibit enhanced inflammation and shorter lifespans, and MANF heterozygous mice exhibit inflammatory phenotypes in various tissues, as well as progressive liver damage, fibrosis, and steatosis. We show that immune cell-derived MANF protects against liver inflammation and fibrosis, while hepatocyte-derived MANF prevents hepatosteatosis. Liver rejuvenation by heterochronic parabiosis in mice further depends on MANF, while MANF supplementation ameliorates several hallmarks of liver aging, prevents hepatosteatosis induced by diet, and improves age-related metabolic dysfunction. Our findings identify MANF as a systemic regulator of homeostasis in young animals, suggesting a therapeutic application for MANF in age-related metabolic diseases.

A new experimental platform facilitates assessment of the transcriptional and chromatin landscapes of aging yeast

Replicative aging of Saccharomyces cerevisiae is an established model system for eukaryotic cellular aging. A limitation in yeast lifespan studies has been the difficulty of separating old cells from young cells in large quantities. We engineered a new platform, the Miniature-chemostat Aging Device (MAD), that enables purification of aged cells at sufficient quantities for genomic and biochemical characterization of aging yeast populations. Using MAD, we measured DNA accessibility and gene expression changes in aging cells. Our data highlight an intimate connection between aging, growth rate, and stress. Stress-independent genes that change with age are highly enriched for targets of the signal recognition particle (SRP). Combining MAD with an improved ATAC-seq method, we find that increasing proteasome activity reduces rDNA instability usually observed in aging cells and, contrary to published findings, provide evidence that global nucleosome occupancy does not change significantly with age.

Simultaneous Profiling of DNA Accessibility and Gene Expression Dynamics with ATAC-Seq and RNA-Seq

This chapter describes sequencing-based methods for profiling dynamic changes in DNA accessibility and gene expression in Saccharomyces cerevisiae. Assay for transposase-accessible chromatin with high-throughput sequencing (ATAC-Seq) is a powerful technique for identifying nucleosome-free regions of the genome. Combining ATAC-Seq with RNA Sequencing (RNA-Seq) is a rapid approach for studying the relationship between genome structure and changes in global patterns of gene expression from a single experiment. A laboratory protocol is presented for these methods as well as examples of typical results and visualizations.

Principles of cellular resource allocation revealed by condition-dependent proteome profiling

Growing cells coordinate protein translation with metabolic rates. Central to this coordination is ribosome production. Ribosomes drive cell growth, but translation of ribosomal proteins competes with production of non-ribosomal proteins. Theory shows that cell growth is maximized when all expressed ribosomes are constantly translating. To examine whether budding yeast function at this limit of full ribosomal usage, we profiled the proteomes of cells growing in different environments. We find that cells produce excess ribosomal proteins, amounting to a constant ≈8% of the proteome. Accordingly, ≈25% of ribosomal proteins expressed in rapidly growing cells does not contribute to translation. Further, this fraction increases as growth rate decreases and these excess ribosomal proteins are employed when translation demands unexpectedly increase. We suggest that steadily growing cells prepare for conditions that demand increased translation by producing excess ribosomes, at the expense of lower steady-state growth rate.

Heterosis as a consequence of regulatory incompatibility

BACKGROUND

The merging of genomes in inter-specific hybrids can result in novel phenotypes, including increased growth rate and biomass yield, a phenomenon known as heterosis. Heterosis is typically viewed as the opposite of hybrid incompatibility. In this view, the superior performance of the hybrid is attributed to heterozygote combinations that compensate for deleterious mutations accumulating in each individual genome, or lead to new, over-dominating interactions with improved performance. Still, only fragmented knowledge is available on genes and processes contributing to heterosis.

RESULTS

We describe a budding yeast hybrid that grows faster than both its parents under different environments. Phenotypically, the hybrid progresses more rapidly through cell cycle checkpoints, relieves the repression of respiration in fast growing conditions, does not slow down its growth when presented with ethanol stress, and shows increased signs of DNA damage. A systematic genetic screen identified hundreds of S. cerevisiae alleles whose deletion reduced growth of the hybrid. These growth-affecting alleles were condition-dependent, and differed greatly from alleles that reduced the growth of the S. cerevisiae parent.

CONCLUSIONS

Our results define a budding yeast hybrid that is perturbed in multiple regulatory processes but still shows a clear growth heterosis. We propose that heterosis results from incompatibilities that perturb regulatory mechanisms, which evolved to protect cells against damage or prepare them for future challenges by limiting cell growth.

Draft Assembly of Elite Inbred Line PH207 Provides Insights into Genomic and Transcriptome Diversity in Maize

Intense artificial selection over the last 100 years has produced elite maize (Zea mays) inbred lines that combine to produce high-yielding hybrids. To further our understanding of how genome and transcriptome variation contribute to the production of high-yielding hybrids, we generated a draft genome assembly of the inbred line PH207 to complement and compare with the existing B73 reference sequence. B73 is a founder of the Stiff Stalk germplasm pool, while PH207 is a founder of Iodent germplasm, both of which have contributed substantially to the production of temperate commercial maize and are combined to make heterotic hybrids. Comparison of these two assemblies revealed over 2500 genes present in only one of the two genotypes and 136 gene families that have undergone extensive expansion or contraction. Transcriptome profiling revealed extensive expression variation, with as many as 10,564 differentially expressed transcripts and 7128 transcripts expressed in only one of the two genotypes in a single tissue. Genotype-specific genes were more likely to have tissue/condition-specific expression and lower transcript abundance. The availability of a high-quality genome assembly for the elite maize inbred PH207 expands our knowledge of the breadth of natural genome and transcriptome variation in elite maize inbred lines across heterotic pools.

Unraveling the message: insights into comparative genomics of the naked mole-rat

Animals have evolved to survive, and even thrive, in different environments. Genetic adaptations may have indirectly created phenotypes that also resulted in a longer lifespan. One example of this phenomenon is the preternaturally long-lived naked mole-rat. This strictly subterranean rodent tolerates hypoxia, hypercapnia, and soil-based toxins. Naked mole-rats also exhibit pronounced resistance to cancer and an attenuated decline of many physiological characteristics that often decline as mammals age. Elucidating mechanisms that give rise to their unique phenotypes will lead to better understanding of subterranean ecophysiology and biology of aging. Comparative genomics could be a useful tool in this regard. Since the publication of a naked mole-rat genome assembly in 2011, analyses of genomic and transcriptomic data have enabled a clearer understanding of mole-rat evolutionary history and suggested molecular pathways (e.g., NRF2-signaling activation and DNA damage repair mechanisms) that may explain the extraordinarily longevity and unique health traits of this species. However, careful scrutiny and re-analysis suggest that some identified features result from incorrect or imprecise annotation and assembly of the naked mole-rat genome: in addition, some of these conclusions (e.g., genes involved in cancer resistance and hairlessness) are rejected when the analysis includes additional, more closely related species. We describe how the combination of better study design, improved genomic sequencing techniques, and new bioinformatic and data analytical tools will improve comparative genomics and ultimately bridge the gap between traditional model and nonmodel organisms.

Systematic identification of cell size regulators in budding yeast

Cell size is determined by a complex interplay between growth and division, involving multiple cellular pathways. To identify systematically processes affecting size control in G1 in budding yeast, we imaged and analyzed the cell cycle of millions of individual cells representing 591 mutants implicated in size control. Quantitative metric distinguished mutants affecting the mechanism of size control from the majority of mutants that have a perturbed size due to indirect effects modulating cell growth. Overall, we identified 17 negative and dozens positive size control regulators, with the negative regulators forming a small network centered on elements of mitotic exit network. Some elements of the translation machinery affected size control with a notable distinction between the deletions of parts of small and large ribosomal subunit: parts of small ribosomal subunit tended to regulate size control, while parts of the large subunit affected cell growth. Analysis of small cells revealed additional size control mechanism that functions in G2/M, complementing the primary size control in G1. Our study provides new insights about size control mechanisms in budding yeast.

Increasing population growth by asymmetric segregation of a limiting resource during cell division

We report that when budding yeast are transferred to low-metal environment, they adopt a proliferation pattern in which division is restricted to the subpopulation of mother cells which were born in rich conditions, before the shift. Mother cells continue to divide multiple times following the shift, generating at each division a single daughter cell, which arrests in G1. The transition to a mother-restricted proliferation pattern is characterized by asymmetric segregation of the vacuole to the mother cell and requires the transcription repressor Whi5. Notably, while deletion of WHI5 alleviates daughter cell division arrest in low-zinc conditions, it results in a lower final population size, as cell division rate becomes progressively slower. Our data suggest a new stress-response strategy, in which the dilution of a limiting cellular resource is prevented by maintaining it within a subset of dividing cells, thereby increasing population growth.

Ndel1-derived peptides modulate bidirectional transport of injected beads in the squid giant axon

Bidirectional transport is a key issue in cellular biology. It requires coordination between microtubule-associated molecular motors that work in opposing directions. The major retrograde and anterograde motors involved in bidirectional transport are cytoplasmic dynein and conventional kinesin, respectively. It is clear that failures in molecular motor activity bear severe consequences, especially in the nervous system. Neuronal migration may be impaired during brain development, and impaired molecular motor activity in the adult is one of the hallmarks of neurodegenerative diseases leading to neuronal cell death. The mechanisms that regulate or coordinate kinesin and dynein activity to generate bidirectional transport of the same cargo are of utmost importance. We examined how Ndel1, a cytoplasmic dynein binding protein, may regulate non-vesicular bidirectional transport. Soluble Ndel1 protein, Ndel1-derived peptides or control proteins were mixed with fluorescent beads, injected into the squid giant axon, and the bead movements were recorded using time-lapse microscopy. Automated tracking allowed for extraction and unbiased analysis of a large data set. Beads moved in both directions with a clear bias to the anterograde direction. Velocities were distributed over a broad range and were typically slower than those associated with fast vesicle transport. Ironically, the main effect of Ndel1 and its derived peptides was an enhancement of anterograde motion. We propose that they may function primarily by inhibition of dynein-dependent resistance, which suggests that both dynein and kinesin motors may remain engaged with microtubules during bidirectional transport.

MRC1-dependent scaling of the budding yeast DNA replication timing program

We describe the DNA replication timing programs of 14 yeast mutants with an extended S phase identified by a novel genome-wide screen. These mutants are associated with the DNA replication machinery, cell-cycle control, and dNTP synthesis and affect different parts of S phase. In 13 of the mutants, origin activation time scales with the duration of S phase. A limited number of origins become inactive in these strains, with inactive origins characterized by small replicons and distributed throughout S phase. In sharp contrast, cells deleted of MRC1, a gene implicated in replication fork stabilization and in the replication checkpoint pathway, maintained wild-type firing times despite over twofold lengthening of S phase. Numerous dormant origins were activated in this mutant. Our data suggest that most perturbations that lengthen S phase affect the entire program of replication timing, rather than a specific subset of origins, maintaining the relative order of origin firing time and delaying firing with relative proportions. Mrc1 emerges as a regulator of this robustness of the replication program.

Cystic papillary hidradenoma of the vulva: case report and review of the literature

Papillary hidradenoma of the vulva is a rare, benign neoplasm arising from apocrine sweat glands of the skin. Frequently, this lesion has been mistaken for carcinoma. The treatment of choice is local excision. The prognosis for patients with this tumor is excellent. We present a patient who is unique because she had a lesion which was the largest ever recorded, and which existed over twice as long as any previously described. This case is also presented to remind the clinician that, despite the gross appearance of the tumor which resembles carcinoma on sectioning, biopsy and histological diagnosis should guide the ultimate management of patients with such lesions. The findings in our patient support the view that no matter how large or how long in existence, hidradenoma remains benign.

Penile premalignant fibroepithelioma of Pinkus

A 63-year-old man with a premalignant fibroepithelioma of Pinkus on the base of his penis is presented. Penile premalignant fibroepitheliomas have not, to the best of our knowledge, been previously described. Because fibroepithelioma of Pinkus is a morphological variant of a basal cell epithelioma, we believe that clinically benign fibroma-like lesions appearing on the penis should be biopsied to rule out the presence of this premalignant lesion.