Unique Molecular Identifiers Research Articles

Ultra-Deep Sequencing As a Predictive Tool for MDS Progression after Allogeneic Hematopoietic Cell Transplant

Background:Allogeneic hematopoietic cell transplantation (alloHCT) is the only curative treatment for patients with myelodysplastic syndromes (MDS). Up to 40% of MDS patients will die within the first year post-alloHCT due to progressive disease or treatment-related mortality. Predictors of post-alloHCT progression are needed. While recent data have shown that gene mutations in pre-transplant samples, including TP53, are associated with worse outcomes post-alloHCT, it is not known whether persistence of patient-specific mutations after alloHCT are predictive of progression. In this study we identified somatic mutations in MDS patients prior to alloHCT and then used ultra-deep sequencing of day 30 and day 100 post-alloHCT bone marrows to determine whether the extent of mutation clearance was associated with outcome.Methods:We selected 55 MDS patients who underwent alloHCT with a mean age of 55.6 years (range 23-71) and IPSS-R scores of very low (1), low (12), intermediate (6), high (23), and very high (13). Transplants included myeloablative and non-myeloablative conditioning regimens with related and unrelated donors. Of the 55 patients, 19 relapsed post-alloHCT and 36 did not with a mean of 242 days to relapse (range 27-1015) and a mean follow-up of 883 days (range 103-2106), respectively. Initial pre-alloHCT bone marrows and paired normal skin (as a source of control DNA) were enhanced exome sequenced (EES) to yield higher coverage in known recurrently mutated AML/MDS genes. To provide high sensitivity at post-alloHCT time points, an ultra-sensitive, error-corrected sequencing approach using unique molecular identifiers (UMIs) was combined with high-coverage depths. Probes were designed to target mutations detected by EES in the pre-alloHCT samples. Initial diagnostic marrow, skin, and day 30 post-alloHCT marrow samples were sequenced in all patients and day 100 post-alloHCT marrows were available in 37 patients. Variant allele fractions (VAFs) of somatic single nucleotide variant (SNV) mutations at day 30 and day 100 post-alloHCT were detected using a custom analysis pipeline employing read-level UMI-based error correction and base-level background error rate corrections. SNV calls were correlated with post-alloHCT progression-free survival and overall survival using Kaplan-Meier and Cox proportional hazards models, respectively.Results:EES was performed to a mean of 230x depth (1310x for recurrently mutated genes). A mean of 40.3 (range 17-151) somatic variants per case were identified with a mean VAF of 29.6%. Of these, a mean of 23 (range 1-70) SNVs per case were targeted and analyzed. Day 30 post-alloHCT marrows were sequenced to a mean depth of 23,062x corresponding to 2,753x unique coverage; day 100 samples were sequenced to 27,936x corresponding to 3,430x unique coverage. A SNV was included in the analysis if the error-corrected pre-transplant VAF was >10% and had >100x unique coverage at day 30. Nineteen of 55 (35%) and 18 of 37 (49%) cases had no mutations detected on post-alloHCT days 30 and 100, respectively. For patients with detectable mutations, the mean VAF per patient at day 30 was 2.35% (range of detected mutations 0.001-25.81%) and 3.44% (range 0.001-21.45%) at day 100. For patients with detectable mutations, the mean maximum VAF per patient at day 30 was 7.21% (range 0.020-88.19%) and 3.44% (range 0.023-38.98%) at day 100.The presence of at least 1 mutation with a VAF greater than 0.5% (equivalent to 1 heterozygous mutant cell in 100) at day 30 post-alloHCT was associated with a worse progression-free survival (median 3.8 vs. 21.9 months; p=0.031). Similar results were obtained using VAF cutoffs of 1% and 2.5% (p= 0.018, p=0.0002, respectively). In contrast, day 100 post-alloHCT marrows did not show a similar correlation. The effect of persistent mutations on overall survival was assessed with a Cox proportional hazards model. A higher mean VAF at day 30 post-alloHCT was associated with worse overall survival (p=0.028); for each mean VAF increase of 1% at day 30, the risk of death increased by about 10% [HR 1.10, 95% CI (1.01,1.20)].Conclusions:Detection of persistent mutations in the bone marrow by ultra-sensitive, error-corrected sequencing at day 30 post-alloHCT is associated with an increased risk of progression. These findings suggest that sequencing could be used as an individualized risk-assessment marker in MDS patients who have undergone alloHCT. DisclosuresUy:Novartis: Consultancy, Other: Travel Suppport; Boehringer Ingelheim: Consultancy; GlycoMimetics: Consultancy.

IFN-γ Induces Distinct Single Cell Transcriptomes in Clinical-Grade Mesenchymal Stromal Cells

INTRODUCTION: Bone marrow derived mesenchymal stromal cells (BMSC) have been widely used to treat or prevent acute graft-versus-host disease, however the variable efficacies reported between clinical trials suggests considerable heterogeneity between clinical-grade BMSC products. Preincubation of BMSC with IFN-γ may improve their immunosuppressive qualities, reduce the functional diversity of the cell product, and decrease the variability in the clinical results obtained. We aimed to identify the distinct BMSC transcriptome pattern induced by IFN-γ and explored BMSC heterogeneity by studying transcriptional diversity at the single cell level.METHODS: Human clinical grade BMSC (n=3) were produced at the GMP Clinical Cell Processing Laboratory in the Department of Transfusion Medicine, NIH under an IRB-approved protocol (NCT01071577). BMSC at passage 3, were cultured with or without IFN-γ (10 µg/mL) and were harvested after 12 hours. Single cell 3' digital gene expression profiling for BMSC from three paired conditions (IFN-γ +/-) were performed via the10xTM GemCodeTM Single Cell Platform (10x Genomics, Pleasanton, CA USA) at a targeted cell number of 5000 cells. Single cell RNA-seq library generations, GEM-RT, and cDNA amplification were performed according to manufacturer's instruction and libraries were sequenced on Illumina Hiseq 3000. The Cell Ranger Single-Cell Software was used for sample demultiplexing, barcode processing, single-cell gene counting, and cell clustering.RESULTS: BMSC samples were analyzed for single cell transcriptome at an average of 5188 cells per condition with a mean number of reads of 50,789 per cell and a median of 4,482 genes per cell. Cluster analysis of the six pooled samples revealed clinical grade BMSCs comprised a heterogenous population of at least 17 clusters. Mean count of unique molecular identifiers (UMI) confirmed that all clusters ubiquitously expressed BMSC phenotype markers (CD105, CD73, and CD90). Predominant clusters in control BMSC (clusters 2, 5, and 6) were enriched with distinct expression profiles, indicative of the diversity and heterogeneity of BMSC. Based on Gene Ontology gene set enrichment analyses, cluster 2 was enriched for genes involved in the extracellular matrix, endoplasmic reticulum lumen, and cell adhesion with high expression of IL1β, a proinflammatory cytokine. Clusters 5 and 6 made up populations enriched with a similar set of genes contributing to the mitotic cell cycle checkpoint system, protein serine-threonine kinase activity, and cell division suggesting proliferation potency. Cluster 6 was further characterized by genes involved in DNA replication and ATP binding. IFN-ɣ dramatically altered cluster arrangement, influencing the cell population and its function. For example, major clusters in control BMSC (clusters 2, 5, 6, and 16) significantly diminished in population size and, instead, new clusters 1, 9, and 15 appeared with IFN-ɣ stimulation (Figure). These IFN-ɣ inducible clusters were enriched for genes regulating chemotaxis, leukocyte migration, and MHC class I and II in addition to the type I interferon signaling pathway. Cluster 1 also expressed IDO1, a heme enzyme producing gene responsible for exhibiting immunosuppressive activity. Cluster 9 possessed distinctive features of genes involved in positive regulation of cAMP-mediated signaling and transport vesicle membranes, suggesting extracellular vesicle and exosome activity. Irrespective of IFN-ɣ stimulation, stagnant clusters enriched with G-protein coupled receptor signaling pathway genes remained, representing a vital population in stem cell self-renewal function.CONCLUSION: Single cell RNA-seq analysis of clinical grade BMSCs revealed a significant heterogeneous population composed of multiple clusters with distinct gene expression patterns. IFN-γ further modified the landscape of transcriptomes of BMSC at a single cell level. This transcriptome diversity may explain the differences in the quality of BMSC products for clinical application. Ongoing experiments include determining the dynamics of the population structure responsible for BMSC diversity and correlating the patterns of single cell transcriptome with the clinical potency of BMSC products. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

NPM1 Mutated AML Can Relapse with Wildtype NPM1 and Evolve with Novel Mutations: Implications for Pathobiology and MRD Monitoring

Introduction: Minimal residual disease (MRD) monitoring by qPCR is feasible in AML with mutated NPM1 (NPM1 mut ) . NPM1 mut leukemia is associated with a favorable outcome and MRD negativity is associated with an excellent prognosis. Cytogenetic and molecular analysis have proven the concept of clonal evolution in relapsed or refractory AML. Data is conflicting on NPM1 mut AML relapsing with wildtype NPM1 (NPM1 negative). In this study we analyzed 105 relapsed NPM1 mut AML patients including 15/105 patients that relapsed with NPM1 negative AML.Aims: To analyze in patients with NPM1 mut AML the frequency and mutational landscape of NPM1 negative AML at relapse and provide insight into clonal evolution.Patients and Methods: Between 2006 und 2017 we investigated a total of 105 intensively treated patients with NPM1 mut AML who attained a complete molecular remission (CMR) and finally relapsed. MRD was monitored by qPCR at 1394 consecutive time points (median=12 per patient, range 3-42). 15/105 patients experienced a NPM1 negative relapse. Unique molecular identifier based sequencing of 63 genes associated with hematological malignancies was performed for patients with NPM1 negative relapse in paired samples at diagnosis and at clinical relapse. The median coverage was 4,900x. Digital error suppression was applied to follow mutations in up to 1/500 molecules (VAF 0.2%).Results: For 15 patients with an NPM1 negative relapse the median time to relapse was 39 months (range 4-66). In contrast for 90/105 patients with an NPM1 mut relapse the median time to relapse was significantly shorter with 14 months (range 3-77, p=0.005). We analyzed the overall survival (OS) following relapse. For patients with NPM1 negative relapse OS at 2 years was 48% [95% CI, 12-77] versus 68% [52-79] for patients with NPM1 mut relapse (n.s.).For 61/90 patients with NPM1 mut relapse the diagnostic sample was available for analysis with the 63 gene panel. The mutational landscape of these patients revealed recurrent co-mutations in 14 different genes in 54/61 patients, with the published pattern: DNMT3A, TET2, FLT3- ITD, NRAS, SRSF2, IDH1, IDH2 and FLT3- TKD were affected in more than 10% of cases. A median of 2 genes were co-mutated per patient, no co-occurring mutation was found in only 7/61 patients (11%).For 12/15 cases with NPM1 negative relapse both diagnostic and relapse samples were available for 63 gene panel sequencing. At diagnosis we identified a total of 33 co-mutations in addition to NPM1 in a total of 12 different genes in 12/12 patients (DNMT3A (n=9 ), TET2 (6), IDH1 (4), SRSF2 (3), PTPN11 (2), IDH2 (1), KRAS (1), NRAS (1), EZH2, (1), STAG2 (1), ASXL1 (1) and FLT3- TKD (1)). All patients had at least one co-mutation with a median of 2 co-mutations per patient at diagnosis (range 1-6). At relapse we identified a total of 37 co-mutations in 13 different genes in 11/12 patients with a median of 2 mutations per patient (range 0-6). Eleven mutations in 9 genes (TET2, STAG2, PTPN11, SRSF2, KRAS, DNMT3A, EZH2, FLT3- TKD and NRAS) ) were lost at relapse and 15 mutations in 10 genes (TP53, RUNX1, NRAS, NF1, IDH2, SRSF2, WT1, IDH1, EZH2, TET2) were gained at relapse. We used digital error correction to reassess these mutations in the diagnostic sample. With a detection limit of up to 0.2% VAF (1/500 molecules) none of these mutations was detectable at diagnosis. DNMT3A was the most prevalent mutation (9/12 at diagnosis, 7/12 at relapse) and compared to the cohort with NPM1 mut relapse the presence of a DNMT3A mutation was associated with NPM1 negative relapse (p=0.024).One patient (8%) relapsed with the respective ancestral clone from diagnosis (w/o NPM1), 9 patients (75%) showed a relapse with the known ancestral clone plus evolution, and 2 patients (17%) relapsed with a completely unrelated clone.Conclusion: We identified 14% of NPM 1mut AML patients who relapsed with an NPM1 negative AML following intensive therapy. NPM1 negative relapse occurs significantly later than NPM1 mut relapse and is associated with the ancestral clone in the majority of patients. NPM1 negativity at relapse evades MRD assessment, however mutations in the ancestral clone should be detectable by NGS with digital error correction. This could be included in an extended MRD monitoring strategy, and performed in addition to qPCR for NPM1 at the routine time points. DisclosuresHöllein:MLL Munich Leukemia Laboratory: Employment. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Fasan:MLL Munich Leukemia Laboratory: Employment. Dicker:MLL Munich Leukemia Laboratory: Employment. Jeromin:MLL Munich Leukemia Laboratory: Employment. Nadarajah:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

MiRge3.0: a comprehensive microRNA and tRF sequencing analysis pipeline.

MicroRNAs and tRFs are classes of small non-coding RNAs, known for their roles in translational regulation of genes. Advances in next-generation sequencing (NGS) have enabled high-throughput small RNA-seq studies, which require robust alignment pipelines. Our laboratory previously developed miRge and miRge2.0, as flexible tools to process sequencing data for annotation of miRNAs and other small-RNA species and further predict novel miRNAs using a support vector machine approach. Although miRge2.0 is a leading analysis tool in terms of speed with unique quantifying and annotation features, it has a few limitations. We present miRge3.0 that provides additional features along with compatibility to newer versions of Cutadapt and Python. The revisions of the tool include the ability to process Unique Molecular Identifiers (UMIs) to account for PCR duplicates while quantifying miRNAs in the datasets, correct erroneous single base substitutions in miRNAs with miREC and an accurate mirGFF3 formatted isomiR tool. miRge3.0 also has speed improvements benchmarked to miRge2.0, Chimira and sRNAbench. Finally, miRge3.0 output integrates into other packages for a streamlined analysis process and provides a cross-platform Graphical User Interface (GUI). In conclusion miRge3.0 is our third generation small RNA-seq aligner with improvements in speed, versatility and functionality over earlier iterations.

Normalization by distributional resampling of high throughput single-cell RNA-sequencing data.

MotivationNormalization to remove technical or experimental artifacts is critical in the analysis of single-cell RNA-sequencing experiments, even those for which unique molecular identifiers are available. The majority of methods for normalizing single-cell RNA-sequencing data adjust average expression for library size (LS), allowing the variance and other properties of the gene-specific expression distribution to be non-constant in LS. This often results in reduced power and increased false discoveries in downstream analyses, a problem which is exacerbated by the high proportion of zeros present in most datasets.ResultsTo address this, we present Dino, a normalization method based on a flexible negative-binomial mixture model of gene expression. As demonstrated in both simulated and case study datasets, by normalizing the entire gene expression distribution, Dino is robust to shallow sequencing, sample heterogeneity and varying zero proportions, leading to improved performance in downstream analyses in a number of settings.Availability and implementationThe R package, Dino, is available on GitHub at https://github.com/JBrownBiostat/Dino. The Dino package is further archived and freely available on Zenodo at https://doi.org/10.5281/zenodo.4897558.Supplementary information Supplementary data are available at Bioinformatics online.

Ultraplex: A rapid, flexible, all-in-one fastq demultiplexer

Background: The first step of virtually all next generation sequencing analysis involves the splitting of the raw sequencing data into separate files using sample-specific barcodes, a process known as “demultiplexing”. However, we found that existing software for this purpose was either too inflexible or too computationally intensive for fast, streamlined processing of raw, single end fastq files containing combinatorial barcodes. Results: Here, we introduce a fast and uniquely flexible demultiplexer, named Ultraplex, which splits a raw FASTQ file containing barcodes either at a single end or at both 5’ and 3’ ends of reads, trims the sequencing adaptors and low-quality bases, and moves unique molecular identifiers (UMIs) into the read header, allowing subsequent removal of PCR duplicates. Ultraplex is able to perform such single or combinatorial demultiplexing on both single- and paired-end sequencing data, and can process an entire Illumina HiSeq lane, consisting of nearly 500 million reads, in less than 20 minutes. Conclusions: Ultraplex greatly reduces computational burden and pipeline complexity for the demultiplexing of complex sequencing libraries, such as those produced by various CLIP and ribosome profiling protocols, and is also very user friendly, enabling streamlined, robust data processing. Ultraplex is available on PyPi and Conda and via Github.

BUTTERFLY: addressing the pooled amplification paradox with unique molecular identifiers in single-cell RNA-seq

The incorporation of unique molecular identifiers (UMIs) in single-cell RNA-seq assays makes possible the identification of duplicated molecules, thereby facilitating the counting of distinct molecules from sequenced reads. However, we show that the naïve removal of duplicates can lead to a bias due to a “pooled amplification paradox,” and we propose an improved quantification method based on unseen species modeling. Our correction called BUTTERFLY uses a zero truncated negative binomial estimator implemented in the kallisto bustools workflow. We demonstrate its efficacy across cell types and genes and show that in some cases it can invert the relative abundance of genes.

Ultraplex: A rapid, flexible, all-in-one fastq demultiplexer.

Background: The first step of virtually all next generation sequencing analysis involves the splitting of the raw sequencing data into separate files using sample-specific barcodes, a process known as "demultiplexing". However, we found that existing software for this purpose was either too inflexible or too computationally intensive for fast, streamlined processing of raw, single end fastq files containing combinatorial barcodes. Results: Here, we introduce a fast and uniquely flexible demultiplexer, named Ultraplex, which splits a raw FASTQ file containing barcodes either at a single end or at both 5' and 3' ends of reads, trims the sequencing adaptors and low-quality bases, and moves unique molecular identifiers (UMIs) into the read header, allowing subsequent removal of PCR duplicates. Ultraplex is able to perform such single or combinatorial demultiplexing on both single- and paired-end sequencing data, and can process an entire Illumina HiSeq lane, consisting of nearly 500 million reads, in less than 20 minutes. Conclusions: Ultraplex greatly reduces computational burden and pipeline complexity for the demultiplexing of complex sequencing libraries, such as those produced by various CLIP and ribosome profiling protocols, and is also very user friendly, enabling streamlined, robust data processing. Ultraplex is available on PyPi and Conda and via Github.

Integrated SMRT Technology with UMI RNA-Seq Reveals the Hub Genes in Stamen Petalody in Camellia oleifera

Male sterility caused by stamen petalody is a key factor for a low fruit set rate and a low yield of Camellia oleifera but can serve as a useful genetic tool because it eliminates the need for artificial emasculation. However, its molecular regulation mechanism still remains unclear. In this study, transcriptome was sequenced and analyzed on two types of bud materials, stamen petalody mutants and normal materials, at six stages of stamen development based on integrated single-molecule real-time (SMRT) technology with unique molecular identifiers (UMI) and RNA-seq technology to identify the hub genes responsible for stamen petalody in C. oleifera. The results show that a large number of alternative splicing events were identified in the transcriptome. A co-expression network analysis of MADSs and all the differentially expressed genes between the mutant stamens and the normal materials showed that four MADS transcription factor genes, CoSEP3.1, CoAGL6, CoSEP3.2, and CoAP3, were predicted to be the hub genes responsible for stamen petalody. Among these four, the expression patterns of CoAGL6 and CoSEP3.2 were consistently high in the mutant samples, but relatively low in the normal samples at six stages, while the patterns of CoSEP3.1 and CoAP3 were initially low in mutants and then were upregulated during development but remained relatively high in the normal materials. Furthermore, the genes with high connectivity to the hub genes showed significantly different expression patterns between the mutant stamens and the normal materials at different stages. qRT-PCR results showed a similar expression pattern of the hub genes in the RNA-seq. These results lay a solid foundation for the directive breeding of C. oleifera varieties and provide references for the genetic breeding of ornamental Camellia varieties.

UMIc: A Preprocessing Method for UMI Deduplication and Reads Correction.

A recent refinement in high-throughput sequencing involves the incorporation of unique molecular identifiers (UMIs), which are random oligonucleotide barcodes, on the library preparation steps. A UMI adds a unique identity to different DNA/RNA input molecules through polymerase chain reaction (PCR) amplification, thus reducing bias of this step. Here, we propose an alignment free framework serving as a preprocessing step of fastq files, called UMIc, for deduplication and correction of reads building consensus sequences from each UMI. Our approach takes into account the frequency and the Phred quality of nucleotides and the distances between the UMIs and the actual sequences. We have tested the tool using different scenarios of UMI-tagged library data, having in mind the aspect of a wide application. UMIc is an open-source tool implemented in R and is freely available from https://github.com/BiodataAnalysisGroup/UMIc.

A unique molecular identifier-based and clonal hematopoiesis-aware approach for accurate mutation calling in cell-free DNA assays.

e22515 Background: Given peripheral blood cells (PBCs) matched cell-free DNA (cfDNA), accurate mutation calling in next generation sequencing (NGS)-based assays relies on discriminating artifacts and clonal hematopoiesis mutations from tissue originated somatic mutations. Although clonal hematopoiesis has been considered in previous overall error modeling, it has not been adapted to PBCs without using unique molecular identifiers (UMIs). Moreover, previous studies on background error profiling were mainly based on healthy controls without matched PBC gDNA, which may lead to potential overestimation of the error rates on those sites with clonal hematopoiesis mutations. Additionally, the fraction of tissue cells is also an important influencing factor but is usually ignored. Methods: We performed UMI-assisted capture-based DNA assays on cfDNA samples, matched PBCs, and oral epithelium cells from 150 healthy donors. A site-specific and subtype-specific background error model was first built for PBCs using the SNVs called from PBCs with matched oral epithelium cells to exclude potential clonal hematopoiesis mutations. Then a similar background model was established for cfDNA with the SNVs inferred from cfDNA to exclude clonal hematopoiesis. The SNVs identified in cfDNA and matched PBCs were separately filtered with the cfDNA and PBC background error models. In this study, we used the ultrasensitive liquid biopsy approach to evaluate paired with tissue and blood samples from 56 early-stage NSCLC patients. All samples are sequenced using NGS target-capture panels covering 29 genes. Results: The mutations were detected in 91.1% of tissue and 67.9% were discovered in plasma. The coincidence rate between tissue and plasma of the 56 early-stage NSCLC patients was 66.1%. Conclusions: We have developed a novel method tailored for UMI-assisted capture-based targeted DNA assays, which outperforms currently available methods in terms of modeling background errors and filtering clonal hematopoietic mutations.

UMI4Cats: an R package to analyze chromatin contact profiles obtained by UMI-4C.

MotivationUMI-4C, a technique that combines chromosome conformation capture (4C) and unique molecular identifiers (UMI), is widely used to profile and quantitatively compare targeted chromosomal contact profiles. The analysis of UMI-4C experiments presents several computational challenges, including the removal of the PCR duplication bias and the identification of differential chromatin contacts.ResultsWe have developed UMI4Cats (UMI-4C Analysis Turned Simple), an R package that facilitates processing, analyzing and visualizing of data obtained by UMI-4C experiments.Availability and implementationUMI4Cats is implemented as an R package supported on Linux, MacOS and MS Windows. UMI4Cats is available from Bioconductor (https://www.bioconductor.org/packages/release/bioc/html/UMI4Cats.html) and GitHub (https://github.com/Pasquali-lab/UMI4Cats).Supplementary information Supplementary data are available at Bioinformatics online.

New Transcriptional Insights into Silent and Active Corticotroph Pituitary Tumors at Single Cell Resolution

Silent pituitary corticotroph tumors derive from the Tpit (aka TBX19) pituitary lineage. Accounting for ~ 30% of corticotroph tumors, they are not infrequently clinically aggressive and invade locally into adjacent sellar structures, making complete surgical resection challenging and contributing to their higher recurrence rates. How silent and active corticotroph tumor subtypes differ is not clear although some studies reported that silent corticotroph tumors exhibit reduced PC1 expression causing impaired POMC processing. We used single cell RNAseq to compare the transcriptome between silent (n = 2) and active (n = 4) corticotroph tumors at the single cell level. We obtained an average of 265 million reads, and 24,682 genes per patient with an average of 1,240 genes expressed and 3,5442 unique molecular identifiers (UMIs) detected per cell. We further defined 5 distinct cell populations from a total of 23,269 cells, namely tumor cells (62%), stromal cells (25%), immune cells (7%), progenitor cells (5%), and a minor population of erythrocytes (1%). Tumor cells clustered in an origin-dependent manner and all expressed POMC and TBX19. However, the gene signatures of silent and active corticotroph tumors differed in 3 major aspects. Firstly, and supporting prior studies, a series of hormone processing peptidase genes, including PC1 (aka PCSK1), PDIA3, SEC11C, SPCS1 and CTSB, were reduced in silent corticotroph tumors (p=5.54e-5) compared to active corticotroph tumors. Secondly, genes involved in organization of secretory vesicles such as SCG5, TIMP1, VGF, SYT17, LGALS3 and CALY were also reduced in silent corticotroph tumors, which could further compound their inability to secrete ACTH. Thirdly, the silent corticotroph tumors exhibited several features of endothelial-to-mesenchymal transition (EMT), including increased expression of the mesenchymal genes CDH2 (aka NCAD), COL1A1, PCDH9, FGF5, ID2 (p=8.4e-3), and loss of EPCAM, which regulate cell migration and movement. Upstream analysis suggested that aberrant STAT3 activation may be related to these changes. Consequently, we noted that the stromal content was higher in silent corticotroph tumors (47.5% vs. 18.13%), concordant with the observed EMT de-differentiation of tumor cells. In summary, our scRNAseq analysis provides an unprecedented precise investigation of the transcriptomic features of thousands of heterogenous corticotroph tumor cells simultaneously. We demonstrate that although silent corticotroph tumor cells still express the pituitary lineage markers PITX1 and TBX19, they exhibit EMT, potentially affording increased migratory capacity at the cost of reduced neuroendocrine function with inability to produce and secrete mature ACTH. Our findings provide novel insights into the pathogenesis of silent versus active corticotroph tumor, but may reveal novel molecular targets for treatment of these challenging tumors.

Epididymal epithelium propels early sexual transmission of Zika virus in the absence of interferon signaling

Recognition of Zika virus (ZIKV) sexual transmission (ST) among humans challenges our understanding of the maintenance of mosquito-borne viruses in nature. Here we dissected the relative contributions of the components of male reproductive system (MRS) during early male-to-female ZIKV transmission by utilizing mice with altered antiviral responses, in which ZIKV is provided an equal opportunity to be seeded in the MRS tissues. Using microRNA-targeted ZIKV clones engineered to abolish viral infectivity to different parts of the MRS or a library of ZIKV genomes with unique molecular identifiers, we pinpoint epithelial cells of the epididymis (rather than cells of the testis, vas deferens, prostate, or seminal vesicles) as a most likely source of the sexually transmitted ZIKV genomes during the early (most productive) phase of ZIKV shedding into the semen. Incorporation of this mechanistic knowledge into the development of a live-attenuated ZIKV vaccine restricts its ST potential.

Bayesian inference of gene expression states from single-cell RNA-seq data.

Despite substantial progress in single-cell RNA-seq (scRNA-seq) data analysis methods, there is still little agreement on how to best normalize such data. Starting from the basic requirements that inferred expression states should correct for both biological and measurement sampling noise and that changes in expression should be measured in terms of fold changes, we here derive a Bayesian normalization procedure called Sanity (SAmpling-Noise-corrected Inference of Transcription activitY) from first principles. Sanity estimates expression values and associated error bars directly from raw unique molecular identifier (UMI) counts without any tunable parameters. Using simulated and real scRNA-seq datasets, we show that Sanity outperforms other normalization methods on downstream tasks, such as finding nearest-neighbor cells and clustering cells into subtypes. Moreover, we show that by systematically overestimating the expression variability of genes with low expression and by introducing spurious correlations through mapping the data to a lower-dimensional representation, other methods yield severely distorted pictures of the data.

Single cell transcriptome sequencing on the Nanopore platform with ScNapBar.

The current ecosystem of single-cell RNA-seq platforms is rapidly expanding, but robust solutions for single-cell and single-molecule full-length RNA sequencing are virtually absent. A high-throughput solution that covers all aspects is necessary to study the complex life of mRNA on the single-cell level. The Nanopore platform offers long read sequencing and can be integrated with the popular single-cell sequencing method on the 10× Chromium platform. However, the high error-rate of Nanopore reads poses a challenge in downstream processing (e.g., for cell barcode assignment). We propose a solution to this particular problem by using a hybrid sequencing approach on Nanopore and Illumina platforms. Our software ScNapBar enables cell barcode assignment with high accuracy, especially if sequencing saturation is low. ScNapBar uses unique molecular identifier (UMI) or Naïve Bayes probabilistic approaches in the barcode assignment, depending on the available Illumina sequencing depth. We have benchmarked the two approaches on simulated and real Nanopore data sets. We further applied ScNapBar to pools of cells with an active or a silenced nonsense-mediated RNA decay pathway. Our Nanopore read assignment distinguishes the respective cell populations and reveals characteristic nonsense-mediated mRNA decay events depending on cell status.

Planarian stem cells specify fate yet retain potency during the cell cycle.

Planarian whole-body regeneration is enabled by stem cells called neoblasts. At least some neoblasts are individually pluripotent. Neoblasts are also heterogeneous, with subpopulations of specialized neoblasts having different specified fates. Fate specification in neoblasts is regulated by fate-specific transcription factor (FSTF) expression. Here, we find that FSTF expression is common in neoblast S/G2/M cell-cycle phases but less common in G1. We find that specialized neoblasts can divide to produce progeny with asymmetric cell fates, suggesting that they could retain pluripotency. Furthermore, no known neoblast class was present in all neoblast colonies, suggesting that pluripotency is not the exclusive property of any known class. We tested this possibility with single-cell transplantations, which indicate that at least some specialized neoblasts are likely clonogenic. On the basis of these findings, we propose a model for neoblast pluripotency in which neoblasts can undergo specialization during the cell cycle without loss of potency.

MiRMaster 2.0: multi-species non-coding RNA sequencing analyses at scale.

Analyzing all features of small non-coding RNA sequencing data can be demanding and challenging. To facilitate this process, we developed miRMaster. After the analysis of over 125 000 human samples and 1.5 trillion human small RNA reads over 4 years, we present miRMaster 2 with a wide range of updates and new features. We extended our reference data sets so that miRMaster 2 now supports the analysis of eight species (e.g. human, mouse, chicken, dog, cow) and 10 non-coding RNA classes (e.g. microRNAs, piRNAs, tRNAs, rRNAs, circRNAs). We also incorporated new downstream analysis modules such as batch effect analysis or sample embeddings using UMAP, and updated annotation data bases included by default (miRBase, Ensembl, GtRNAdb). To accommodate the increasing popularity of single cell small-RNA sequencing data, we incorporated a module for unique molecular identifier (UMI) processing. Further, the output tables and graphics have been improved based on user feedback and new output formats that emerged in the community are now supported (e.g. miRGFF3). Finally, we integrated differential expression analysis with the miRNA enrichment analysis tool miEAA. miRMaster is freely available at https://www.ccb.uni-saarland.de/mirmaster2.

Estimation of tuna population by the improved analytical pipeline of unique molecular identifier-assisted HaCeD-Seq (haplotype count from eDNA)

Many studies have investigated the ability to identify species from environmental DNA (eDNA). However, even when individual species are identified, the accurate estimation of their abundances by traditional eDNA analyses has been still difficult. We previously developed a novel analytical method called HaCeD-Seq (Haplotype Count from eDNA), which focuses on the mitochondrial D-loop sequence. The D-loop is a rapidly evolving sequence and has been used to estimate the abundance of eel species in breeding water. In the current study, we have further improved this method by applying unique molecular identifier (UMI) tags, which eliminate the PCR and sequencing errors and extend the detection range by an order of magnitude. Based on this improved HaCeD-Seq pipeline, we computed the abundance of Pacific bluefin tuna (Thunnus orientalis) in aquarium tanks at the Tokyo Sea Life Park (Kasai, Tokyo, Japan). This tuna species is commercially important but is at high risk of resource depletion. With the developed UMI tag method, 90 out of 96 haplotypes (94%) were successfully detected from Pacific bluefin tuna eDNA. By contrast, only 29 out of 96 haplotypes (30%) were detected when UMI tags were not used. Our findings indicate the potential for conducting non-invasive fish stock surveys by sampling eDNA.

Non-linear Normalization for Non-UMI Single Cell RNA-Seq.

Single cell RNA-seq data, like data from other sequencing technology, contain systematic technical noise. Such noise results from a combined effect of unequal efficiencies in the capturing and counting of mRNA molecules, such as extraction/amplification efficiency and sequencing depth. We show that such technical effects are not only cell-specific, but also affect genes differently, thus a simple cell-wise size factor adjustment may not be sufficient. We present a non-linear normalization approach that provides a cell- and gene-specific normalization factor for each gene in each cell. We show that the proposed normalization method (implemented in “SC2P" package) reduces more technical variation than competing methods, without reducing biological variation. When technical effects such as sequencing depths are not balanced between cell populations, SC2P normalization also removes the bias due to uneven technical noise. This method is applicable to scRNA-seq experiments that do not use unique molecular identifier (UMI) thus retain amplification biases.