How mispriming events could be creating artifacts in your library prep (and what you can do to prevent it)

Next-generation sequencing technology has been advancing at an incredibly rapid rate; what started as only genome sequencing now encompasses an incredible amount of RNA sequencing techniques as well. These range from standard RNA-seq, to miRNA-seq, Ribo-seq, to HITS-CLIP (high-throughput sequencing of RNA isolated by crosslinking immunoprecipiation). While these technological advances are now widely used (and have been invaluable to the scientific community), they are not fully mature technologies and we are still learning about potential artifacts that may arise and how to combat them; mispriming events are a significant and under-studied contributor to errors in sequencing data.

What is a mispriming event?

Reverse transcription is an important part of any RNA-sequencing technique. The RNA in question is first converted into cDNA, which is then PCR amplified and converted in a library from there (there are various methods for library preparation, depending on what kind of technique you are using). However, the conversion of RNA into cDNA by reverse transcriptase requires a DNA primer to start the process. This primer is complementary to the RNA, binding to it and allowing for reverse transcription to take place. A mispriming event is when this process occurs at a place where the DNA primer is not perfectly complementary to the RNA.

Two recent papers have highlighted how reverse transcription mispriming events can have a considerable impact on the library preparation process and result in error. Gurp, McIntyre and Verhoeven [1] conducted an RNA-seq experiment focusing on reads that mapped to ERCC spikes (artificial and known RNA fragments that are added to RNA-seq experiments as a control). Because the sequence of these ERCC spikes is already known, detecting mismatches in the sequences is relatively straightforward.

Their findings were striking: they found that 1) RNA-to-DNA mispriming events were the leading cause of deviations from the true sequence (as opposed to DNA-to-DNA mispriming events that can occur later on in the library preparation process), and 2) these mispriming events are non-random and indeed show specific and predictable patterns. For example, if the first nucleotide of an RNA-seq read starts with A or T, rA-dC and rU-dC mispriming events are common. In positions 2 – 6, rU-dG and rG-dT are also quite common, which lines up with the observation that these are the most stable mismatched pairs [2]. Needless to say, these kind of mispriming events can cause huge issues for various type of downstream analysis, particularly identification of SNPs and RNA-editing sites; eliminating these biases will be extremely important for future experiments (Figure 1). 

journal.pone.0085583.g002

Figure 1: Common base mismatches and their locations [1]

As of right now, we do not have good, sophisticated methods of eliminating these types of mispriming events from our datasets. Eliminating the first 10 bases of reads will solve the problem, but will also involve throwing out real data with the artifacts. Given the fact that these mispriming events do follow predictable patterns, it is possible that in the future, we could devise programs to identify and correct mispriming events, or even modify hexamer design to exclude ones that result in frequent mispriming.

Frustratingly, mispriming events can occur even when the priming oligo is quite lengthy. HITS-CLIP has been greatly important in discovering many protein-RNA interactions [3]; however, a recent paper published by Gillen et al. [4]  demonstrated that mispriming events even with a long DNA primer can create a significant artifact, creating read pileups that align to the genomic occurrences of the adaptor sequence, making it appear as though there are protein-RNA interactions occurring at that locus.

Part of HITS-CLIP library preparation involves attachment of a 3’ RNA adaptor to the protein bound RNA. A DNA oligo perfectly complementary to this RNA sequence serves as the primer for conversion of this RNA into cDNA, and it is this DNA oligo that leads to significant mispriming events. Although the DNA primer is long enough to be extremely specific, sequences that are complementary to only the last 6 nucleotides of the primer are still enough to result in a mispriming event, which converts alternative RNAs into cDNAs that eventually get amplified in the library.

Gillen et al. analyzed 44 experiments from 17 research groups, and showed that the adaptor sequence was overrepresented by 1.5-fold on average–and sometimes as high as 6-fold (Figure 2)!

12864_2016_2675_Fig1_HTML

Figure 2: Over-representation of DNA primer sequences can be found in multiple datasets from different groups, indicating the possibility of a widespread problem. 

And since only 6 complementary nucleotides are needed to result in a mispriming event, how can we eliminate this artifactual data?

Gillen et al. devised an ingenious and simple method of reducing this artifact by using a nested reverse transcription primer (Figure 3). By ‘nested primer’, they are referring to a primer that is not perfectly complementary to the 3’ adaptor, but rather stops 3 nucleotides short of being fully flush with the adaptor. This, combined with a full-length PCR primer (that is, flush with the adaptor sequence) with a ‘protected’ final 3 nucleotides (note: in this instance, ‘protected’ mean usage of phosphorothioate bonds in the final 3 oligo bases, which prevents degradation by exonucleases. Without this protective bond, the mispriming artifact is simply shifted downstream 3 bases), is enough to almost completely eliminate mispriming artifacts. This allows for significantly improved library quality and increased sensitivity!

12864_2016_2675_Fig2_HTML

Figure 3: A nested reverse transcription primer combined with a protected PCR primer can eliminate sequencing artifacts almost entirely. 

Although we have been working with sequencing technologies for many years now, we still have a lot to discover about hidden artifacts in the data. It’s becoming increasingly important to stay aware of emerging discoveries of these biases and make sure we are doing everything we can to eliminate this from our data.

Have you ever had an experience with sequencing artifacts in your data? Tell us in the comments!

International biological material shipment information for various countries

Many scientific researchers prefer to outsource their next generation sequencing projects to commercial service providers to get access to the latest instruments and scientific expertise.

However, there are some countries in the world that do not allow the export of biological samples (tissue samples, DNA, RNA etc.) or require several formal agreements and multi-level clearance.

In this post, we’ll highlight some general information about shipping samples out of several major countries, primarily to the US. Some of this is based on our experience working with many international researchers who use Genohub to outsource their sequencing.

China

China, for example, does not allow the import or export of biological samples, as confirmed by multiple courier service agents1. Major Chinese service providers require biological samples to be shipped to their Hong Kong address to avoid delay or loss of samples2,3.

In a rare situation, a Chinese group of researchers was able to ship DNA samples to the US using FedEx. They have also detailed their experience and have some advice regarding sample shipment that can be potentially useful to other groups willing to do the same4.

Brazil

To export biological material from Brazil, several documents such as Material Transfer Agreement and Institutional invoice of specimen exported, are required for customs clearance. A detailed cover letter in both Portuguese and English that can help Customs officials in Brazil (IBAMA) and the USA (USFWS) properly assess the authorization to export and import specimens is also required5. It could take several weeks to obtain these documents so researchers need to plan their work in advance.

India

Until 2016, The Indian Council of Medical Research made decisions on shipment of biological samples on a case-by-case basis6. However, these regulations have since been lifted since August 2016 and researchers have to follow several guidelines for biological materials to qualify for transport to foreign countries for research purposes7.

According to a FedEx India employee, a non-infectious certificate from an authentic laboratory and a detailed description of the included biological samples is sufficient for customs clearance from India. Any pathogenic material is not allowed to be shipped internationally.

Europe

We haven’t come across any issues shipping samples from European countries and generally, a properly declared biological shipment can be exported without any hassles.

The current Universal Postal Union regulations for shipping biological material have been comprehensively summarized in an official document. This document also lists the countries that allow or ban the import/export of biological substances8.

Please consult our shipping guide for more details on how to prepare your shipment to ship samples to USA – https://genohub.com/dna-rna-shipping-for-ngs/#USA.

If you know of any countries that require a lot of formal paperwork for export of biological substances for research or sequencing purposes, feel free to comment below. I’ll update the blog with this information.

References:

(1)     China Country Snapshot https://smallbusiness.fedex.com/international/country-snapshots/china.html.

(2)     Sample Preparation; Shipping – Novogene https://en.novogene.com/support/sample-preparation/.

(3)     Sample submission guidelines – BGI http://www.bgisample.com/yangbenjianyi/BGI-TS-03-12-01-001 Suggestions for Sample Delivery(NGS) B0.pdf.

(4)     Community/ZJU-China Letter about Shipping DNA – 2015.igem.org http://2015.igem.org/Community/ZJU-China_Letter_about_Shipping_DNA.

(5)     Shipping and Customs http://symbiont.ansp.org/ixingu/shipping/index.html.

(6)    Centre removes ICMR approval for import/export of human biological samples http://www.dnaindia.com/india/report-centre-removes-icmr-approval-for-importexport-of-human-biological-samples-2245910.

(7)     Indian Council of Medical Research http://icmr.nic.in/ihd/ihd.htm.

(8)     WFCC Regulations http://www.wfcc.info/pdf/wfcc_regulations.pdf

Sequencing trends in early 2017

Every month, ~5,000 unique queries for sequencing are submitted using Genohub’s NGS project matching engine: https://genohub.com/ngs/. Briefly, a user chooses the NGS application they are interested in (e.g. exome, RNA-Seq), the number of reads or coverage they’d like to achieve and the number of samples they plan on sequencing. Genohub’s matching engine, takes this input calculates the sequencing output required to meet the desired coverage and recommends services, filterable by sequencing instrument, read length, and library preparation kit. Results can be sorted by price, turnaround time and selected for immediate ordering.

Every query that’s submitted is recorded giving us a unique perspective into what types of NGS services researchers are actually interested in.

DNA-Seq

First, it’s important to note that DNA-seq is our default option in the matching engine: https://genohub.com/ngs/. Due to this bias, you can’t really compare it to other services being ordered so it’s a good idea to just throw away this data point. Of DNA-seq services that are actually ordered, this breaks down into: whole human genome sequencing, re-sequencing, and metagenomics sequencing. The most frequently used instruments for this service are currently the HiSeq X, HiSeq 3000/4000 and NextSeq. With PacBio’s release of the Sequel, requests have significantly increased this quarter compared to PacBio service requests in the last 4 quarters. We expect this trend to continue through 2017.

RNA-Seq

The pie chart above breaks down the types of RNA-seq services requested in the first three months of 2017. Total RNA-seq represents all applications where rRNA is depleted prior to library preparation, whereas mRNA-seq represents all applications where mRNA is enriched. In 2016, the number of Total RNA-seq projects was half that of this year. We attribute this to a growing interest in non-coding RNA and the availability of higher throughput sequencing runs. As sequencing costs drop and rRNA depletion becomes more affordable, researchers are asking for more biological information.  Today, the Nextseq and HiSeq 3000/4000 are the most commonly used instruments for any RNA-seq application. Counting applications continue to dominate, although requests for de novo transcriptome alignments are steady rising over the previous year. Whereas in the past, 1×50 and 1×75 were the most frequently requested read length for RNA counting applications, around 2x more researchers are requesting paired-end sequencing versus last year.

Methylation analysis

Compared to last year, there is an increased interest in WGBS as compared to RRBS and MeDIP. With the advent of the HiSeq X and it’s compatibility with WGBS applications, more researchers are finding whole genome based applications easier and more informative than reduced representation bisulfite sequencing.

Instrument trends

By far the biggest trend this year was the number of long read requests on the PacBio Sequel. Whereas in the past, Mate-pair library prep was more popular, we’re starting to see this service decline, and long read sequencing be ordered more frequently. Hybrid Ilumina/PacBio reads are also being more frequently ordered to improve the quality of assemblies. Long-reads are being requested to detect functional elements in human genomes that are missed by short-read sequencing. We should add that requests for 10X Genomics services have started to increase, although they are too small right now to make any meaningful comments. We currently don’t have providers offering Oxford Nanopore services on Genohub, so can’t comment here either.

This month NovaSeq services are expected to be available on Genohub. We expect there to be a lag phase as kinks are worked out, before this becomes a popular instrument request.

The future

Having spent the last 4 years receiving sequencing requests and performing consultation, it’s clear that new technology does influence behavior. With reduced sequencing costs, we see clients not only including more control and duplicates, but also looking at RNA-seq from a more global perspective, and beginning to become more interested in long reads. Clients that previously only performed exome-seq are now turning to whole genome sequencing on the HiSeq X. Researchers that normally only look at coding RNA’s are starting to show interest in long non-coding and small RNAs. Overall, faster and cheaper sequencing does tend to promote better science. Gone are the n=1 days of sequencing.

Beginner’s Handbook to High Throughput Sequencing

book-311432_640

As sequencing becomes more ubiquitous, we find researchers struggling with concepts like ‘paired-end’, designing a custom sequencing primer, cluster density, and technical library prep details, like why can’t small RNA and mRNA both be prepared in the same library and sequenced? This is partially the fault of industry, e.g. are 100M ‘paired-end reads’ comprised of 200M, 100M or 50M single reads [We like to denote this as 100M paired end reads (50M reads in each direction)], and partially due to all the moving parts: new sequencing and library prep chemistries, technology jargon and complexities in data analysis.

Seeing first time researchers struggle (on hundreds of sequencing projects), we sought to put together a guide to help the sequencing novice get a strong foothold on starting a sequencing project. This guide is called our Beginner’s Handbook to Next Generation Sequencing.

The guide is broken up into four main sections:

  1. Sequencing instruments and design of a sequencing project
  2. Library prep
  3. Sample isolation
  4. Providers we recommend you contact for analyzing your data

Whether you are new to NGS or an experienced NGS user, we recommend you check it out and ask questions. We’ll be updating the guide on a regular basis, so if you have recommendations, please post them here. Thanks!

 

 

RNA-Seq considerations when working with nucleic acid derived from FFPE

RNA-seq from FFPE samples

Millions of formalin-fixed paraffin-embedded (FFPE) tissue sections are stored in oncology tissue banks and pathology laboratories around the world. Formalin fixation followed by embedding paraffin has historically been a popular preservation method in histological studies as morphological features of the original tissue remain intact. However for RNA-seq or other gene expression methods, formalin fixation and paraffin embedding can degrade and modify RNA, complicating retrospective analysis using this commonly used archival method.

During the fixation and embedding process RNA is affected in the following ways:

  1. Degradation of RNA to short ~100 base fragments as a result of sample treatment during fixation or long term storage in paraffin.
  2. Formaldehyde modification of RNA. Formaldehyde modification can block base pairing and can cause cross-linking to other macromolecules. These RNA modifications include hydroxymethyl and methylene bridge cross-links on amine moieties of adenine bases.
  3. High variability in the degree of RNA degradation and modification in FFPE samples precludes transcriptomic similarity and gene expression correlation studies, or simply forces researchers to exclude certain samples.
  4. Oligo-dT approaches are not recommended when amplifying RNA as most RNA fragments derived from FFPE no long contain a poly(A) tail making rRNA depletion a necessary first step prior to RNA-seq.

If formalin fixation and paraffin embedding can’t be avoided, Ahlfen et al., nicely summarize best practices for improving RNA quality and yield from FFPE samples. These include:

  1. Starting fixation and cutting samples into thin pieces to avoid tissue autolysis.
  2. Reduction of fixation time (< 24 hours) to reduce irreversible cross-linking and RNA fragmentation during storage of FFPE blocks.
  3. Utilizing a method to reverse cross-linking during RNA isolation. These include heating RNA to remove some formaldehyde cross-linking. Reaction of formaldehyde with amino groups in bases and proteins are largely irreversible and inhibit cDNA synthesis.
  4. Use of a rRNA depletion step and random priming as opposed to oligo-dT based reversed transcription.
  5. RNA QC methods such as a measurement of RNA integrity or one of several RT-PCR based kits to qualify a sample prior to RNA-seq.

Despite these challenges, FFPE samples are frequently used in transcriptomic studies and in many cases correlate nicely with fresh frozen samples (Hedegaard et al., 2014; Li et al., 2014; Zhao et al., 2014). The study of somatic mutations continues to remain a challenge in FFPE tissue due to fragmentation and the presence of artifacts. Nevertheless, RNA molecules from FFPE are being used regularly for investigating both non-coding and coding parts of the genome.

If you have FFPE blocks or total RNA and would like to perform gene expression analysis by RNA-Seq, we recommend you start with a NGS service provider who has specific experience with FFPE RNA isolation, QC, library preparation, sequencing and data analysis. Providers with this experience can be found using this search on Genohub: https://genohub.com/ngs/?r=mt3789#q=4c5f2d036f.

 

Accurate measurement of error rate and base quality in Illumina sequencing runs

With new instrumentation, cluster chemistries, software updates and continuously updated library preparation reagents; accurately monitoring sequencing run quality has become increasing difficult.  In a recent paper by Manley et al., 2016, the authors develop an open source tool called the Percent Perfect Reads (PPR) plot to monitor base quality.

PPR uses PhiX alignment and calculates percent of reads with 0–4 mismatches.  A PPR plot contains a cycle-by-cycle representation of the percentage of reads with mismatches. PPR was originally introduced with the original Genome Analyzer and retired in 2014.

PPR is developed as an alternative to the Phred-like Q score for determining run quality and has the following advantages:

  1. PPR is independently calculated, unlike Illumina’s Q Score which is calculated with instrument dependent variables (vary by instrument, chemistry, software)
  2. PPR is a direct measure of error unlike Q score’s which rely on a table of data, generated under ideal sequencing circumstances
  3. Q scores tend to overestimate quality
  4. Unlike with Q scores, PPR allows the user to identify the source of sequencing error

By examining a PPR profile, the following issues are distinguishable:

  1. Adapter read through (sequencing cycles are longer than the library insert and the run reads through the adapter sequence)
  2. Repetitive or low diversity sequences
  3. Imaging problems
  4. Over/under clustering
  5. Chemistry problems (cluster reagents are not working properly)

The PPR plot program is compatible with HiSeq 2000/2500, NextSeq 500, and MiSeq instruments. It’s written in Perl and R, and accepts FASTQ files as input. The PPR software package is available at http://openwetware.org/wiki/BioMicroCenter:PPR_Program (BioMicro Center, Massachusetts Institute of Technology, Cambridge, MA, USA).

 

Illumina unveils NovaSeq 5000 and 6000

Illumina NovaSeq

Today, at the annual J.P. Morgan Healthcare Conference, Illumina announced the release of a new series of instruments called NovaSeq. Continuing the use of ExAmp cluster amplification and patterned nano-wells that form the basis of HiSeq 3000/4000 HiSeq X Ten and HiSeq X Five flow cell technology, Illumina further reduced the spacing between nanowells to increase cluster density and data output. In the end, this promises to produce ~ 2-3x more reads than a single 8 lane HiSeq X flow cell.

Here are the specs available on day 1 of launch:

Number of instruments being launched: 2; NovaSeq 5000 and 6000

Non-technical application based restrictions: No, unlike the HiSeq X Ten or HiSeq X Five; these instruments will not have application based restrictions. Illumina plans to continue restricting HiSeq X instruments to WGS applications (1).

Potential technical based restrictions: Notable is the absence of Nextera based DNA or Nextera Exome in the list of compatible library preparation kits. Mate-pair based Nextera kits are however listed as compatible (2). This may indicate there are template (library) size restrictions on this instrument (similar to HiSeq 3000/4000 and HiSeq X).

Instrument availability: NovaSeq 6000 will begin shipping in March 2017 and NovaSeq 5000 will begin shipping mid-2017.

Anticipated availability on GenohubIn April 2017, researchers will be able to order NovaSeq based sequencing. This hinges on on-time instrument delivery to our partnering service providers.

Instrument cost: NovaSeq 5000 and 6000 Systems are priced at $850,000 and $985,000 respectively

Target Market: Research labs that cannot afford the capital cost of a HiSeq X Five or HiSeq X Ten System and don’t want to deal with the restrictions. HiSeq X Five and Ten systems are restricted from running RNA-seq or exome based libraries.

Other updates: RFID added to make sure loading is done properly, reduction in the number of steps in a sequencing workflow (from 38 to 8) (1) and flow cell loading is automated.

Cbot or onboard clustering: onboard

Tunable output: 4 flow cells are available. NovaSeq S1 and S2 flow cells are compatible with both NoveSeq 5000 and 6000 systems while NovaSeq S3 and S4 are exclusive to NovaSeq 6000 instruments.

Two color or Four color chemistry: Two color, like the NextSeq 500

Number of lanes: S1 and S2 have two lanes whereas S3 and S4 have four lanes

Available read lengths: 2×50, 2×100 and 2×150

Run times: < 19, 29 and 40 hours for 2×50, 2×100 and 2×150 bp read lengths respectively

Output: 

Instrument and flow cell Reads per flow cell *(billion) Output from 2×150 bp run (Gb) *
NovaSeq 5000/6000 S1 1.6 500
NovaSeq 5000/6000 S2 3.3 1000
NovaSeq 6000 S3 6.6 2000
NovaSeq 6000 S4 10 3000

*Output and read numbers based on a single flow cell

Number of flow cells that can be run at once: 1 or 2 flow cells can be run on both the NovaSeq 5000 or 6000

So what does this mean for the sequencing industry? Clearly the Novaseq was launched to target research labs that can’t afford the capital costs of the HiSeq X series but want to upgrade from their current HiSeq instruments. NovaSeq S3 and S4 flow cells promise to produce 2-3x more reads than a single 8 lane HiSeq X flow cell (2.6-3 billion reads).  Of course,  if NovaSeq is priced to run 2-3x more expensive than a HiSeq X flow cell, the cost it takes to sequence a genome will be the same. When reagent pricing is available, this will be more clear.

2016 was a tough year for Illumina as it lost one third of its value. As Illumina launches another instrument geared for the research market, much continues to hinge on federally funded research grants to fuel growth. A focus on developing clinical based applications, insurance reimbursable tests and a global shift toward diagnostics is going to be required for sustained growth. ‘Market generation’ activities, as were initiatives like Helix and Grail are steps in this direction.