Pan-cancer subclonal mutation analysis of 7,827 tumors predicts clinical outcome.

Intra-tumor heterogeneity is an important driver of tumor evolution and therapy response. Advances in precision cancer treatment will require understanding of mutation clonality and subclonal architecture. Currently the slow computational speed of subclonal reconstruction hinders large cohort studies. To overcome this bottleneck, we developed Clonal structure identification through Pairwise Penalization, or CliPP, which clusters subclonal mutations using a regularized likelihood model. CliPP reliably processed whole-genome and whole-exome sequencing data from over 12,000 tumor samples within 24 hours, thus enabling large-scale downstream association analyses between subclonal structures and clinical outcomes. Through a pan-cancer investigation of 7,827 tumors from 32 cancer types, we found that high subclonal mutational load (sML), a measure of latency time in tumor evolution, was significantly associated with better patient outcomes in 16 cancer types with low to moderate tumor mutation burden (TMB). In a cohort of prostate cancer patients participating in an immunotherapy clinical trial, high sML was indicative of favorable response to immune checkpoint blockade. This comprehensive study using CliPP underscores sML as a key feature of cancer. sML may be essential for linking mutation dynamics with immunotherapy response in the large population of non-high TMB cancers.

Accelerated somatic mutation calling for whole-genome and whole-exome sequencing data from heterogenous tumor samples.

Accurate detection of somatic mutations in DNA sequencing data is a fundamental prerequisite for cancer research. Previous analytical challenges were overcome by consensus mutation calling from four to five popular callers. This, however, increases the already nontrivial computing time from individual callers. Here, we launch MuSE 2, powered by multistep parallelization and efficient memory allocation, to resolve the computing time bottleneck. MuSE 2 speeds up 50 times more than MuSE 1 and eight to 80 times more than other popular callers. Our benchmark study suggests combining MuSE 2 and the recently accelerated Strelka2 achieves high efficiency and accuracy in analyzing large cancer genomic data sets.

DeMixSC: a deconvolution framework that uses single-cell sequencing plus a small benchmark dataset for improved analysis of cell-type ratios in complex tissue samples.

Bulk deconvolution with single-cell/nucleus RNA-seq data is critical for understanding heterogeneity in complex biological samples, yet the technological discrepancy across sequencing platforms limits deconvolution accuracy. To address this, we introduce an experimental design to match inter-platform biological signals, hence revealing the technological discrepancy, and then develop a deconvolution framework called DeMixSC using the better-matched, i.e., benchmark, data. Built upon a novel weighted nonnegative least-squares framework, DeMixSC identifies and adjusts genes with high technological discrepancy and aligns the benchmark data with large patient cohorts of matched-tissue-type for large-scale deconvolution. Our results using a benchmark dataset of healthy retinas suggest much-improved deconvolution accuracy. Further analysis of a cohort of 453 patients with age-related macular degeneration supports the broad applicability of DeMixSC. Our findings reveal the impact of technological discrepancy on deconvolution performance and underscore the importance of a well-matched dataset to resolve this challenge. The developed DeMixSC framework is generally applicable for deconvolving large cohorts of disease tissues, and potentially cancer.

Accelerated somatic mutation calling for whole-genome and whole-exome sequencing data from heterogenous tumor samples.

Accurate detection of somatic mutations in DNA sequencing data is a fundamental prerequisite for cancer research. Previous analytical challenge was overcome by consensus mutation calling from four to five popular callers. This, however, increases the already nontrivial computing time from individual callers. Here, we launch MuSE2.0, powered by multi-step parallelization and efficient memory allocation, to resolve the computing time bottleneck. MuSE2.0 speeds up 50 times than MuSE1.0 and 8-80 times than other popular callers. Our benchmark study suggests combining MuSE2.0 and the recently expedited Strelka2 can achieve high efficiency and accuracy in analyzing large cancer genomic datasets.

Estimation of tumor cell total mRNA expression in 15 cancer types predicts disease progression.

Single-cell RNA sequencing studies have suggested that total mRNA content correlates with tumor phenotypes. Technical and analytical challenges, however, have so far impeded at-scale pan-cancer examination of total mRNA content. Here we present a method to quantify tumor-specific total mRNA expression (TmS) from bulk sequencing data, taking into account tumor transcript proportion, purity and ploidy, which are estimated through transcriptomic/genomic deconvolution. We estimate and validate TmS in 6,590 patient tumors across 15 cancer types, identifying significant inter-tumor variability. Across cancers, high TmS is associated with increased risk of disease progression and death. TmS is influenced by cancer-specific patterns of gene alteration and intra-tumor genetic heterogeneity as well as by pan-cancer trends in metabolic dysregulation. Taken together, our results indicate that measuring cell-type-specific total mRNA expression in tumor cells predicts tumor phenotypes and clinical outcomes.

MuSE: A Novel Approach to Mutation Calling with Sample-Specific Error Modeling.

Accurate detection of somatic mutations in genetically heterogeneous tumor cell populations using next-generation sequencing remains challenging. We have developed MuSE, Mutation calling using a Markov Substitution model for Evolution, a novel approach for modeling the evolution of the allelic composition of tumor and normal tissue at each reference base. It adopts a sample-specific error model to depict inter-tumor heterogeneity, which greatly improves the overall accuracy. Here, we describe the method and provide a tutorial on the installation and application of MuSE.

Novel Anaplastic Thyroid Cancer PDXs and Cell Lines: Expanding Preclinical Models of Genetic Diversity.

CONTEXT: Anaplastic thyroid cancer (ATC) is a rare, aggressive, and deadly disease. Robust preclinical thyroid cancer models are needed to adequately develop and study novel therapeutic agents. Patient-derived xenograft (PDX) models may resemble patient tumors by recapitulating key genetic alterations and gene expression patterns, making them excellent preclinical models for drug response evaluation. OBJECTIVE: We developed distinct ATC PDX models concurrently with cell lines and characterized them in vitro and in vivo. METHODS: Fresh thyroid tumor from patients with a preoperative diagnosis of ATC was surgically collected and divided for concurrent cell line and PDX model development. Cell lines were created by generating single cells through enzymatic digestion. PDX models were developed following direct subcutaneous implantation of fresh tumor on the flank of immune compromised/athymic mice. RESULTS: Six ATC PDX models and 4 cell lines were developed with distinct genetic profiles. Mutational characterization showed one BRAF/TP53/CDKN2A, one BRAF/CDKN2A, one BRAF/TP53, one TP53 only, one TERT-promoter/HRAS, and one TERT-promoter/KRAS/TP53/NF2/NFE2L2 mutated phenotype. Hematoxylin-eosin staining comparing the PDX models to the original patient surgical specimens show remarkable resemblance, while immunohistochemistry stains for important biomarkers were in full concordance (cytokeratin, TTF-1, PAX8, BRAF). Short tandem repeats DNA fingerprinting analysis of all PDX models and cell lines showed strong concordance with the original tumor. PDX successful establishment rate was 32%. CONCLUSION: We have developed and characterized 6 novel ATC PDX models with 4 matching cell lines. Each PDX model harbors a distinct genetic profile, making them excellent tools for preclinical therapeutic trials.

Structure of SPH (self-incompatibility protein homologue) proteins: a widespread family of small, highly stable, secreted proteins.

SPH (self-incompatibility protein homologue) proteins are a large family of small, disulfide-bonded, secreted proteins, initially found in the self-incompatibility response in the field poppy (Papaver rhoeas), but now known to be widely distributed in plants, many containing multiple members of this protein family. Using the Origami strain of Escherichia coli, we expressed one member of this family, SPH15 from Arabidopsis thaliana, as a folded thioredoxin fusion protein and purified it from the cytosol. The fusion protein was cleaved and characterised by analytical ultracentrifugation, circular dichroism and nuclear magnetic resonance (NMR) spectroscopy. This showed that SPH15 is monomeric and temperature stable, with a ß-sandwich structure. The four strands in each sheet have the same topology as the unrelated proteins: human transthyretin, bacterial TssJ and pneumolysin, with no discernible sequence similarity. The NMR-derived structure was compared with a de novo model, made using a new deep learning algorithm based on co-evolution/correlated mutations, DeepCDPred, validating the method. The DeepCDPred de novo method and homology modelling to SPH15 were then both used to derive models of the 3D structure of the three known PrsS proteins from P. rhoeas, which have only 15-18% sequence homology to SPH15. The DeepCDPred method gave models with lower discreet optimised protein energy scores than the homology models. Three loops at one end of the poppy structures are postulated to interact with their respective pollen receptors to instigate programmed cell death in pollen tubes.

DeepCDpred: Inter-residue distance and contact prediction for improved prediction of protein structure.

Rapid, accurate prediction of protein structure from amino acid sequence would accelerate fields as diverse as drug discovery, synthetic biology and disease diagnosis. Massively improved prediction of protein structures has been driven by improving the prediction of the amino acid residues that contact in their 3D structure. For an average globular protein, around 92% of all residue pairs are non-contacting, therefore accurate prediction of only a small percentage of inter-amino acid distances could increase the number of constraints to guide structure determination. We have trained deep neural networks to predict inter-residue contacts and distances. Distances are predicted with an accuracy better than most contact prediction techniques. Addition of distance constraints improved de novo structure predictions for test sets of 158 protein structures, as compared to using the best contact prediction methods alone. Importantly, usage of distance predictions allows the selection of better models from the structure pool without a need for an external model assessment tool. The results also indicate how the accuracy of distance prediction methods might be improved further.

Kissing and nanotunneling mediate intermitochondrial communication in the heart.

Mitochondria in many types of cells are dynamically interconnected through constant fusion and fission, allowing for exchange of mitochondrial contents and repair of damaged mitochondria. However, constrained by the myofibril lattice, the ∼6,000 mitochondria in the adult mammalian cardiomyocyte display little motility, and it is unclear how, if at all, they communicate with each other. By means of target-expressing photoactivatable green fluorescent protein (PAGFP) in the mitochondrial matrix or on the outer mitochondrial membrane, we demonstrated that the local PAGFP signal propagated over the entire population of mitochondria in cardiomyocytes on a time scale of ∼10 h. Two elemental steps of intermitochondrial communications were manifested as either a sudden PAGFP transfer between a pair of adjacent mitochondria (i.e., "kissing") or a dynamic nanotubular tunnel (i.e., "nanotunneling") between nonadjacent mitochondria. The average content transfer index (fractional exchange) was around 0.5; the rate of kissing was 1‰ s(-1) per mitochondrial pair, and that of nanotunneling was about 14 times smaller. Electron microscopy revealed extensive intimate contacts between adjacent mitochondria and elongated nanotubular protrusions, providing a structural basis for the kissing and nanotunneling, respectively. We propose that, through kissing and nanotunneling, the otherwise static mitochondria in a cardiomyocyte form one dynamically continuous network to share content and transfer signals.

Turbo fast three-dimensional carotid artery black-blood MRI by combining three-dimensional MERGE sequence with compressed sensing.

PURPOSE: In this study, we sought to investigate the feasibility of turbo fast three-dimensional (3D) black-blood imaging by combining a 3D motion-sensitizing driven equilibrium rapid gradient echo sequence with compressed sensing. METHODS: A pseudo-centric phase encoding order was developed for compressed sensing-3D motion-sensitizing driven equilibrium rapid gradient echo to suppress flow signal in undersampled 3D k-space. Nine healthy volunteers were recruited for this study. Signal-to-tissue ratio, contrast-to-tissue ratio (CTR) and CTR efficiency (CTReff ) between fully sampled and undersampled images were calculated and compared in seven subjects. Moreover, isotropic high resolution images using different compressed sensing acceleration factors were evaluated in two other subjects. RESULTS: Wall-lumen signal-to-tissue ratio or CTR were comparable between the undersampled and the fully sampled images, while significant improvement of CTReff was achieved in the undersampled images. At an isotropic high spatial resolution of 0.7 × 0.7 × 0.7 mm(3) , all undersampled images exhibited similar level of the flow suppression efficiency and the capability of delineating outer vessel wall boundary and lumen-wall interface, when compared with the fully sampled images. CONCLUSION: The proposed turbo fast compressed sensing 3D black-blood imaging technique improves scan efficiency without sacrificing flow suppression efficiency and vessel wall image quality. It could be a valuable tool for rapid 3D vessel wall imaging.

Automatic segmentation of white matter hyperintensities by an extended FitzHugh & Nagumo reaction diffusion model.

PURPOSE: To evaluate the efficiency and reproducibility of the extended FitzHugh & Nagumo (FHN) reaction-diffusion model proposed in this study for white matter hyperintensities (WMH) segmentation. MATERIALS AND METHODS: Five types of magnetic resonance T2-weighted fluid-attenuated inversion-recovery (T2FLAIR) images of 127 patients with different scanning parameters from five clinical scanner systems were selected for this study. After skull and scalp removal and denoise, the T2FLAIR images were processed by the proposed extended FHN model to obtain WMH. This new technique replaced the global threshold constant with a local threshold matrix. RESULTS: There was no significant difference between the segmentation results of the training set and the manual contouring against those between the test set and the manual contouring based on similarity index (SI) values (P = 0.5217). The SI values of the five types of T2FLAIR images were 86.0% ± 15.4%, 85.8% ± 10.5%, 84.1% ± 14.8%, 87.2% ± 14.6%, 86.3% ± 12.7%, respectively, comparing the segmentation results using the proposed method to the manual delineations. The overall SI value of the images was 86.5% ± 14.5%. This approach also demonstrated a better WMH segmentation performance over its classic form (P < 0.001). CONCLUSION: The proposed approach is efficient and could provide a more effective and convenient tool for clinical quantitative WMH analysis.