Kenneth J. Livak

The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells

Defining the transcriptional dynamics of a temporal process such as cell differentiation is challenging owing to the high variability in gene expression between individual cells. Time-series gene expression analyses of bulk cells have difficulty distinguishing early and late phases of a transcriptional cascade or identifying rare subpopulations of cells, and single-cell proteomic methods rely on a priori knowledge of key distinguishing markers. Here we describe Monocle, an unsupervised algorithm that increases the temporal resolution of transcriptome dynamics using single-cell RNA-Seq data collected at multiple time points. Applied to the differentiation of primary human myoblasts, Monocle revealed switch-like changes in expression of key regulatory factors, sequential waves of gene regulation, and expression of regulators that were not known to act in differentiation. We validated some of these predicted regulators in a loss-of function screen. Monocle can in principle be used to recover single-cell gene expression kinetics from a wide array of cellular processes, including differentiation, proliferation and oncogenic transformation.Single-cell expression profiling by RNA-Seq promises to exploit cell-to-cell variation in gene expression to reveal regulatory circuitry governing cell differentiation and other biological processes. Here, we describe Monocle, a novel unsupervised algorithm for ordering cells by progress through differentiation that dramatically increases temporal resolution of expression measurements in a model of skeletal muscle differentiation. This reordering unmasks switch-like changes in expression of key regulatory factors, reveals sequentially organized waves of gene regulation, and exposes novel regulators of cell differentiation. A loss-of function screen revealed that many of these inhibitors act through regulatory elements also used by pro-myogenic factors to activate downstream genes. This study demonstrates that single-cell expression analysis by Monocle can uncover novel regulatory interactions governing differentiation.

The chromatin structure of specific genes: I. Evidence for higher order domains of defined DNA sequence.

Abstract When the chromatin of Drosophila is examined by digestion with DNAase I or micrococcal nuclease, no general structural organization above the level of the nucleosome is revealed by the cleavage pattern. In contrast, the DNAase I cleavage pattern of specific regions of the Drosophila chromosome shows discrete bands with sizes ranging from a few kilobase pairs (kb) to more than 20 kb. Visualization of such higher order bands was achieved by the use of the Southern blotting technique. The DNAase I-cleaved fragments were transferred onto a nitrocellulose sheet after size fractionation by gel electrophoresis. Hybridization was then carried out with radioactively labeled cloned fragments of DNA from D. melanogaster. For the five different chromosomal regions examined, each gives a unique pattern of higher order bands on the autoradiogram; the patterns are different for different regions. Restriction enzyme cleavage of the fragments generated indicates that the preferential DNAase I cleavage sites in chromatin are position-specific. The chromosomal regions bounded by preferential DNAase I cleavage sites are referred to as supranucleosomal or higher order domains for purposes of discussion and analysis. The micrococcal nuclease cleavage pattern of chromatin at specific loci was also examined. In the one case studied in detail, this nuclease also cleaves at position-specific sites.

Taking qPCR to a higher level: Analysis of CNV reveals the power of high throughput qPCR to enhance quantitative resolution.

This paper assesses the quantitative resolution of qPCR using copy number variation (CNV) as a paradigm. An error model is developed for real-time qPCR data showing how the precision of CNV determination varies with the number of replicates. Using samples with varying numbers of X chromosomes, experimental data demonstrates that real-time qPCR can readily distinguish four copes from five copies, which corresponds to a 1.25-fold difference in relative quantity. Digital PCR is considered as an alternative form of qPCR. For digital PCR, an error model is shown that relates the precision of CNV determination to the number of reaction chambers. The quantitative capability of digital PCR is illustrated with an experiment distinguishing four and five copies of the human gene MRGPRX1. For either real-time qPCR or digital PCR, practical application of these models to achieve enhanced quantitative resolution requires use of a high throughput PCR platform that can simultaneously perform thousands of reactions. Comparing the two methods, real-time qPCR has the advantage of throughput and digital PCR has the advantage of simplicity in terms of the assumptions made for data analysis.

Studies of cloned sequences from four drosophila heat shock loci

DNA cloned from the D. melanogaster (Oregon R) heat shock loci at 63BC and 95D codes for the 83,000 and the 68,000 dalton heat shock proteins, respectively. Both coding sequences occur once per haploid genome. Sequences complementary to messenger RNA for the 70,000 dalton heat shock protein are represented five times, twice at 87A and three times at 87 C. The copies at 87A differ characteristically from those at 87C in an interval of a few hundred bp near the 5 end of the messenger sequence, and the corresponding two classes of hsp 70 messenger RNA are found on polysomes after heat shock. Within this differential region, there is about 15% divergence between messenger sequences cloned from the two loci, while in the rest of the messenger region examined the homology is much closer although still imperfect. Unexpectedly, considerable homology is found between the sequence for the 68,000 dalton heat shock protein at 95D and the sequences for the 70,000 dalton protein at 87A and 87C, and between these sequences and a site in 87D. Messenger RNA molecules of 2.4, 2.55 and 3.05 kb code for the 68,000, 70,000 and 83,000 dalton heat shock proteins and hybridize to apparently uninterrupted DNA sequences of 2.1, 2.25 and 2.6 kb, respectively.

Single-cell gene expression analysis reveals genetic associations masked in whole-tissue experiments

Gene expression in multiple individual cells from a tissue or culture sample varies according to cell-cycle, genetic, epigenetic and stochastic differences between the cells. However, single-cell differences have been largely neglected in the analysis of the functional consequences of genetic variation. Here we measure the expression of 92 genes affected by Wnt signaling in 1,440 single cells from 15 individuals to associate single-nucleotide polymorphisms (SNPs) with gene-expression phenotypes, while accounting for stochastic and cell-cycle differences between cells. We provide evidence that many heritable variations in gene function--such as burst size, burst frequency, cell cycle-specific expression and expression correlation/noise between cells--are masked when expression is averaged over many cells. Our results demonstrate how single-cell analyses provide insights into the mechanistic and network effects of genetic variability, with improved statistical power to model these effects on gene expression.

Single-cell RNA-seq supports a developmental hierarchy in human oligodendroglioma

Although human tumours are shaped by the genetic evolution of cancer cells, evidence also suggests that they display hierarchies related to developmental pathways and epigenetic programs in which cancer stem cells (CSCs) can drive tumour growth and give rise to differentiated progeny. Yet, unbiased evidence for CSCs in solid human malignancies remains elusive. Here we profile 4,347 single cells from six IDH1 or IDH2 mutant human oligodendrogliomas by RNA sequencing (RNA-seq) and reconstruct their developmental programs from genome-wide expression signatures. We infer that most cancer cells are differentiated along two specialized glial programs, whereas a rare subpopulation of cells is undifferentiated and associated with a neural stem cell expression program. Cells with expression signatures for proliferation are highly enriched in this rare subpopulation, consistent with a model in which CSCs are primarily responsible for fuelling the growth of oligodendroglioma in humans. Analysis of copy number variation (CNV) shows that distinct CNV sub-clones within tumours display similar cellular hierarchies, suggesting that the architecture of oligodendroglioma is primarily dictated by developmental programs. Subclonal point mutation analysis supports a similar model, although a full phylogenetic tree would be required to definitively determine the effect of genetic evolution on the inferred hierarchies. Our single-cell analyses provide insight into the cellular architecture of oligodendrogliomas at single-cell resolution and support the cancer stem cell model, with substantial implications for disease management.

CEL-Seq2: sensitive highly-multiplexed single-cell RNA-Seq.

Single-cell transcriptomics requires a method that is sensitive, accurate, and reproducible. Here, we present CEL-Seq2, a modified version of our CEL-Seq method, with threefold higher sensitivity, lower costs, and less hands-on time. We implemented CEL-Seq2 on Fluidigm’s C1 system, providing its first single-cell, on-chip barcoding method, and we detected gene expression changes accompanying the progression through the cell cycle in mouse fibroblast cells. We also compare with Smart-Seq to demonstrate CEL-Seq2’s increased sensitivity relative to other available methods. Collectively, the improvements make CEL-Seq2 uniquely suited to single-cell RNA-Seq analysis in terms of economics, resolution, and ease of use.

Clonal evolution in patients with chronic lymphocytic leukaemia developing resistance to BTK inhibition

Resistance to the Brutons tyrosine kinase (BTK) inhibitor ibrutinib has been attributed solely to mutations in BTK and related pathway molecules. Using whole-exome and deep-targeted sequencing, we dissect evolution of ibrutinib resistance in serial samples from five chronic lymphocytic leukaemia patients. In two patients, we detect BTK-C481S mutation or multiple PLCG2 mutations. The other three patients exhibit an expansion of clones harbouring del(8p) with additional driver mutations (EP300, MLL2 and EIF2A), with one patient developing trans-differentiation into CD19-negative histiocytic sarcoma. Using droplet-microfluidic technology and growth kinetic analyses, we demonstrate the presence of ibrutinib-resistant subclones and estimate subclone size before treatment initiation. Haploinsufficiency of TRAIL-R, a consequence of del(8p), results in TRAIL insensitivity, which may contribute to ibrutinib resistance. These findings demonstrate that the ibrutinib therapy favours selection and expansion of rare subclones already present before ibrutinib treatment, and provide insight into the heterogeneity of genetic changes associated with ibrutinib resistance.

Methods for qPCR gene expression profiling applied to 1440 lymphoblastoid single cells

Highlights ► Microfluidic arrays enable analysis of 96 qPCR assays on 1440 single cells. ► Detailed methods on obtaining qPCR data and performing preliminary data processing. ► Data from sufficient cells to address noise inherent in single-cell transcription. ► Methods used for conventional qPCR do not necessarily apply to single-cell qPCR.

Molecular load of pathologically occult metastases in pelvic lymph nodes is an independent prognostic marker of biochemical failure after localized prostate cancer treatment.

PURPOSEnThirty percent of patients treated with curative intent for localized prostate cancer (PC) experience biochemical recurrence (BCR) with rising serum prostate-specific antigen (sPSA), and of these, approximately 50% succumb to progressive disease. More discriminatory staging procedures are needed to identify occult micrometastases that spawn BCR.nnnPATIENTS AND METHODSnPSA mRNA copies in pathologically normal pelvic lymph nodes (N0-PLN) from 341 localized PC patients were quantified by real-time reverse-transcriptase polymerase chain reaction. Based on comparisons with normal lymph nodes and PLN with metastases and on normalization to 5 x 10(6) glyceraldehyde-3-phosphate dehydrogenase mRNA copies, normalized PSA copies (PSA-N) and a threshold of PSA-N 100 or more were selected for continuous and categorical multivariate analyses of biochemical failure-free survival (BFFS) compared with established risk factors.nnnRESULTSnAt median follow-up of 4 years, the BFFS of patients with PSA-N 100 or more versus PSA-N less than 100 was 55% and 77% (P = .0002), respectively. The effect was greatest for sPSA greater than 20 ng/mL, 25% versus 60% (P = .014), Gleason score 8 or higher, 21% versus 66% (P = .0002), stage T3c, 18% versus 64% (P = .001), and high-risk group (50% v 72%; P = .05). By continuous analysis PSA-N was an independent prognostic marker for BCR (P = .049) with a hazard ratio of 1.25 (95% CI, 1.001 to 1.57). By categorical analysis, PSA-N 100 or more was an independent variable (P = .021) with a relative risk of 1.98 (95% CI, 1.11 to 3.55) for BCR compared with PSA-N less than 100.nnnCONCLUSIONnPSA-N 100 or more is a new, independent molecular staging criterion for localized PC that identifies high-risk group patients with clinically relevant occult micrometastases in N0-PLN, who may benefit from additional therapy to prevent BCR.