WIXLIB

New Technologies for Sensitive,Low-Input RNA-SeqClontech Laboratories, Inc.

OutlineIntroductionSingle-Cell-Capable mRNA-Seq Using SMART Technology SMARTer Ultra Low RNA Kit for the Fluidigm C1 System SMART-Seq v4 Ultra Low Input RNA Kit for SequencingTotal RNA-Seq Applications SMARTer Stranded RNA-Seq Kit SMARTer Stranded Total RNA Sample Prep Kit - HI Mammalian SMARTer Stranded Total RNA-Seq Kit - Pico Input MammalianExpanding Applications for SMART Technology DNA SMART ChIP-Seq Kit2

Next-Gen Sequencing: RNA-SeqcDNA synthesis for a complete representation of the transcriptomeLow-input totalRNA samples Industry standard for low-inputand single-cell RNA-seqSingle cells Full-length gene bodycoverageDegraded,LCM, or FFPEsamplesProkaryoticsamples3SMARTTechnology Simplified workflow Highest sensitivity andreproducibility High-quality sequencinglibraries

Transcriptome Analysis with NGS RNA-seq produces millions of sequences from complex RNA samples. With thispowerful approach, we can:– Measure gene expression/evaluate differential gene expression between differentconditions, cell types, etc.– Discover and annotate complete transcripts– Discover and characterize alternative splicing (isoforms), polyadenylation & SNPs In recent years, more people have been interested in investigating such applications atthe single-cell levelThe Technology and Biology of Single-Cell RNA SequencingKolodziejczyk A., et al. (2015) Molecular Cell 58(4):610–6204

SMARTer cDNA Synthesis for NGSdT Primed mRNA Polyadenylated RNA Single cell capacitySMART-Seq v4 Ultra Low Input RNA Kitfor Sequencing Use with Ion Torrent orIllumina platformsN6 Primed Coding and non-coding RNA Non-polyadenylated RNA Degraded samplesSMARTer Stranded Total RNA-Seq Kit - Pico InputMammalian Use with Illumina platforms Use with highly degraded samples (FFPE)SMARTer Stranded RNA-Seq Kit Use with Illumina platformsSMARTer Ultra Low RNA Kitfor the Fluidigm C1 SystemcDNA SynthesisRiboGone - MammalianSMARTer Stranded Total RNASample Prep- HI Mammalian 96 single cells in parallel5 Use with typical input RNA samples

Importance of Studying Single CellsGenome Research (2005) 15:1388–1392ColdSpring Harbor Laboratory Press Gene expression can vary significantlybetween cells (transcription occurs inbursts). Average expression of a populationmay not necessarily correlate withgene expression at the level of anindividual cell.Fig 1. Histograms showing the expression levels of 96 cells expressing ActB inlogarithmic and linear scale (inset).6

Transcriptome Analysis of IndividualCancer Cells Individual cells can be categorizedaccording to their cell line of originbased on their transcriptome 12 individual cancer cells were isolatedfrom three different cancer cell lines– Four cells each from prostate (PC3and LNCaP) and bladder (T24) celllines Global gene expression profiles wereused to analyze each single-celltranscriptomeFull-length mRNA-seq from single-cell levels of RNA and individual circulating tumor cells.Ramsköld, D., et al. (2012) Nature Biotechnology 30(8):777–782.7

Transcriptome Analysis of Individual ction introa8

Our Solutions for mRNA-SeqdT PrimingUltra-low-input total RNAand 1–1,000 cells mRNA Polyadenylated RNA Single-cell capacity With no strand informationSMART-Seq v4 Ultra Low Input RNA Kit for SequencingInput 10 pg–10 ng; 1–1,000 cells Ultra-low-input total RNA, poly(A ) RNA Single-cell capacity Compatible with Ion Torrent and Illumina platformsSMARTer Ultra Low RNA Kit for the Fluidigm C1 SystemInput 1–1,000 cellsFor the Fluidigm C1cell-capture system9 96 single cells in parallel Compatible with Illumina platforms

Single-Cell and Ultra-Low InputRNA-Seq Solutions—Timeline1996SMARTtechnologyfrom ClontechSMARTerUltra Low RNAKit for IlluminaSequencing2011RNAsequenced1970s10SMARTer UltraLow RNA Kitfor the FluidigmC1 SystemSMARTer UltraLow Input RNAKit for IlluminaSequencing - HV2013SMART-Seq2innovationsSMARTer UltraLow Input RNAKit forSequencing - v32014SMART-Seq v4Ultra Low InputRNA Kit forSequencing2015

SMARTer Ultra Low RNA Kitfor the Fluidigm C1 SystemEnrichLoad &CaptureWash &StainIsolateLyse, RT& AmplifyPrepareLibrarySequenceAnalyze* Slide from Fluidigm11

SMART-Seq v4—Advancements Input 10 pg–10 ng; 1–1,000 cells for mRNA-seq Optimized template-switching oligo: Incorporates LNA and with ourproprietary knowledge of template switching Improved sensitivity and reproducibility More genes identified Higher yield Simplified protocolThe SMART-Seq v4 kit outperforms all previous generationsof SMARTer Ultra Low kits by increasing sensitivity andreproducibility.12

SMART-Seq v4—Technology The SMARTScribe Reverse Transcriptase(RT) makes cDNA. When the SMARTScribe RT reaches the 5' endof the RNA, its terminal transferase activity addsa few nucleotides. The SMART-Seq Oligonucleotide base-pairswith the non-templated nucleotide stretch,creating an extended template to allow theSMARTScribe RT to continue replicating. The SMART-Seq primer and oligo serve asuniversal priming sites for cDNA amplification byPCR.13

Comparing ULv3, SMART-Seq v4, andSMART-Seq2 Reduced background Improved sensitivity andreproducibilityImprovements to thetemplate-switching oligoSequencing Data Comparing cDNA Synthesis Methods10 pg Mouse Brain RNAInputProtocolULv3SMART-Seq v4SMART-Seq2181718Number of PCR cycles4.0 (paired-end)Number of reads 39Mapped to genome969796957293Mapped to exons737376766667Mapped to introns212119202827Mapped to intergenic regions6.06.24.74.75.85.8ReplicateNo. of transcripts identifiedPercentage of reads (%):14

Comparing ULv3, SMART-Seq v4, andSMART-Seq2ULv3R 0.911 0.683FPKMs 100R 0.376 0.19115SMART-Seq v4R 0.972 0.820FPKMs 100R 0.739 0.604SMART-Seq2R 0.966 0.706FPKMs 100R 0.496 0.288

Mapping Statistics from 10 pg–10 ngof MAQC ControlsSequencing Metrics from MAQC and ERCC ControlsHBRRRNA sourceHURRHBRRHURR10 pg10 ngNumber of paired-end reads(Millions)2.33.4Number of PCR cycles188Input amountTranscripts with FPKM 0.115,482 15,338 15,421 17,612 17,756 17,588 25,111 25,218 25,128 25,075 25,118 25,146Transcripts with FPKM 112,598 12,561 12,707 14,491 14,516 14,445 18,346 18,163 18,386 18,061 17,792 17,907Percentage of reads (%):Mapped to rRNA1.21.21.10.70.70.65.65.45.64.04.04.1Mapped to pped to genome929092949494949494969696Mapped to exons777978808081737373777777Mapped to introns181717141414212121181718Mapped to 5.7Duplicates212222202020181818212019Mapped to ERCC3.73.83.81.21.21.12.92.82.90.950.960.9716

Differential Expression Compared toMAQC qPCR Data20y 0.7891x 0.0135R 0.78310 pgqPCR Data (log2)1510554 equencing Data (log2)20y 1.0535x 0.0166R 0.93410 ngqPCR Data (log2)1510841 transcripts50-5-10-15-20-20-15-10-505Sequencing Data (log2)17

Identification of Long Transcripts18

SMARTer cDNA Synthesis for NGSdT Primed mRNA Polyadenylated RNA Single cell capacitySMART-Seq v4 Ultra Low Input RNA Kitfor Sequencing Use with Ion Torrent orIllumina platformsN6 Primed Coding and non-coding RNA Non-polyadenylated RNA Degraded samplesSMARTer Stranded Total RNA-Seq Kit - Pico InputMammalian Use with Illumina platforms Use with highly degraded samples (FFPE)SMARTer Stranded RNA-Seq Kit Use with Illumina platformsSMARTer Ultra Low RNA Kitfor the Fluidigm C1 SystemcDNA SynthesisRiboGone - MammalianSMARTer Stranded Total RNASample Prep- HI Mammalian 96 single cells in parallel19 Use with typical input RNA samples

Our Solutions for Total RNA-SeqN6 Priming Coding and non-coding RNA Degraded, FFPE, and LCM samples Non-polyadenylated RNAWith Strand InformationSMARTer Stranded Total RNA Sample Prep Kit - HI Mammalian(RiboGone - Mammalian kit built in)Input 100 ng–1 g of total RNASMARTer Stranded Total RNA Sample Prep Kit - Low InputMammalian(with RiboGone - Mammalian kit)Input 10 ng–100 ng of total RNASMARTer Stranded Total RNA-Seq Kit - Pico Input MammalianInput 250 pg–10 ng of total RNA20

SMARTer Stranded RNA-Seq with rRNADepletionTwo configurations:Low Input (10 ng–100 ng) RiboGone - Mammalian SMARTerStranded Total RNA Sample PrepHigh Input (100 ng–1 mg) SMARTer Stranded Total RNASample Prep Kit - HI21

SMARTer Stranded RNA-Seq for Low-InputTotal RNA SamplesAnalyses of sequencing dataSequence Alignment Metrics (Input: 10 ng)HumanUniversalHumanBrainNumber of reads (Millions)6.87.7Number of genes identified14,56313,839Mapped to rRNA0.9%0.7%Mapped to mtRNA4.7%2.9%Mapped uniquely to genome76%75%Mapped to exons70%66%Mapped to introns47%49%Mapped to intergenic regions53%51%RNA sourcePercentage of reads (%):22MAQC Analysis

SMARTer Stranded Total RNA Sample PrepKit - HI MammalianMAQC AnalysisAnalyses of sequencing dataSequence Alignment Metrics (Input: 400 ng)RNA sourceNumber of reads (Millions)HumanUniversalHumanBrain8.5 (paired-end)17,57017,600Mapped to rRNA0.3%5.3%Mapped to genome94%88%Mapped uniquely to genome91%84%Mapped to exons43%50%Mapped to introns43%33%Mapped to intergenic regions14%12%99.3%98.8%Number of genes identifiedPercentage of reads (%):ERCC transcriptswith correct strand23ERCC analysis

SMARTer Stranded Total RNA-Seq Kit Pico Input MammalianSMARTer Stranded Total RNA-Seq Kit - Pico Input Mammalian RNA-seq library prep kit directly resulting in Illumina-compatible libraries Input 250 pg–10 ng of mammalian total RNA Incorporates LNA technology, leading to greater sensitivity Streamlined workflow, including the depletion of rRNA in the form of ribosomalcDNA using a novel, proprietary technology Maintains strand information Uses random priming to generate information from coding and non-coding RNA Compatible with a range of RNA qualities (e.g., FFPE & LCM samples)24

Stranded Total RNA-Seq - Pico Inputwith ribosomal cDNA depletion25

Consistent Sequencing Metrics Across a100-Fold Input RangeSequencing Alignment MetricsRNA sourceMouse Brain Total RNAInput amount (ng)Library yield (ng/μl)1010.5114.89.93Number of readsNumber of transcripts FPKM 1Pearson/Spearman correlationsCorrect strand per biol. annotation (%)0.258.36.910.17.485.767.2612,61512,28612,5282.6 million 9.5Mitochondrial8.88.78.38.58.38.47.57.9Overall mapping (%)86.485.986.186.285.184.883.383.1Duplicate rate (%)12.812.517.317.831.328.844.240.2Proportion of total reads (%)26

Stranded Total RNA-Seq - Pico Inputfrom FFPE SamplesSequencing Alignment MetricsRNA sourceInput amount (ng)Human Liver Total RNA - FFPE10Ribosomal cDNAremoval10.25YesNumber of reads0.25No1 million (paired-end)Number of transcriptsFPKM 111,75211,36010,3684,501Number of transcriptsFPKM 0.115,35814,68012,7934,50798.398.198.397.1Correct strand perbiol. annotation (%)Proportion of total reads al4.03.53.11.4Overall mapping (%)85.884.078.596.1Duplicate rate (%)22.639.052.544.027

SMART Technology for DNA Sequencing—Expanding ApplicationsDNA SMART ChIP-Seq Kit For low-input ChIP-seq for Illumina platforms Single-tube workflow; under 4 hours Compatible with dsDNA or ssDNA (100 pg–10 ng) Ligation-free addition of Illumina adapters Generates high-complexity libraries from picogram amounts of input DNA28

Combined ChIP Elute and ChIP-Seq Kits—WorkflowDNA SMART ChIP-Seq Kit29

High-Quality Data from Low-Cell NumberChIP-Seq ExperimentsSequencing Metrics from Specified Numbers of CellsChIP antibodyInput (293T cells)PCR cyclesLibrary yield (nM)Peaks 22,564200,00017,7761,56391% overlap1,000,0002,36050,00017,09988% overlapPercentage of reads (%):Reads mapped92.788.684.375.81,000,000Uniquely mapped reads79.074.870.459.62,682Useful reads(uniquely t rate0.850.850.710.5710,00016,7775,78786% overlap1,000,0003,890ENCODE15,56989% overlap(ENCODE data from U. Washington—293T cells)Analysis performed with 15–18 million reads per sample301,4502,047

Robust Libraries from Low-Cell-NumberChIP-Seq ExperimentsReproducibility is maintained for low cell numbers1,000,000 cells200,000 cells50,000 cells10,000 cells1,000,000 cells(no ab)ENCODERefSeq Genes31