Vedlegg til Forbruk til Packbio sekvensieringssystem
Fil: Application-note-Kinnex-full-length-RNA-kit-for-isoform-sequencing.pdf
Vedlegg til anbudet Forbruk til Packbio sekvensieringssystem fra Universitetet i Oslo.
Page 1
Application note
Kinnex full-length RNA kit
for isoform sequencing
Introduction
Alternative splicing (AS) in eukaryotic species PacBio RNA sequencing has been shown to offer the
generates functional diversity by expressing different longest read lengths and highest quality reads that are
combinations of exons in the same gene. Accurate required for isoform discovery and isoform
characterization of full-length transcript isoforms quantification (Pardo-Palacios et al., 2024). The
generated by AS is critical for biological and disease Kinnex™ full-length RNA kit takes full-length cDNA
studies. Bulk RNA-Seq using short reads cannot fully generated by the Iso-Seq express 2.0 kit as input and
resolve isoform structures, as the complex nature of outputs a sequencing-ready library that results in an 8-
AS prohibits unambiguous transcript assembly with fold throughput increase compared to typical Iso-Seq
even the most sophisticated computational tools (Stark libraries. Combined with the Read Segmentation & Iso-
et al., 2019). Long-read RNA-Seq using PacBio® Seq analysis in SMRT® Link software, PacBio offers
technology (the Iso-Seq® method) eliminates the need cost-effective isoform sequencing that does not require
for transcript assembly by sequencing full-length orthogonal sequencing methods. SMRT Link
cDNAs and enables new discoveries across many software produces an isoform classification report with
applications (Figure 1). abundance information that can be used by tertiary
analysis tools.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 2
al., 2023). The Iso-Seq method has also been used in
plant and animal research to create high-quality
genome annotations (Zhang et al., 2023) as well as
identify parental-specific isoform expressions (Wang et
al., 2020).
Kinnex full-length RNA kit
The Kinnex full-length RNA kit utilizes the MAS-Seq
Figure 1. Long-read RNA sequencing eliminates the need for method to increase throughput on PacBio long-read
transcript assembly, which cannot accurately resolve the sequencers. MAS-Seq is a concatenation method for
isoform structure. Long-read RNA-Seq using PacBio (the Iso- joining cDNA molecules into longer fragments
Seq method) sequences the entire full-length cDNA to (Al’Khafaji et al., 2023). HiFi reads generated from
provide an unambiguous view of the transcriptome. sequencing the concatenated molecules can then be
broken up bioinformatically to retrieve the original
Full-length RNA sequencing with the cDNA sequences. The result is higher throughput and
Iso-Seq method reduced sequencing needs for cost-effective isoform
sequencing.
Traditional RNA-Seq fragments cDNA for short-read
sequencing (100–200 bp) must be followed by The PacBio Iso-Seq workflow in SMRT Link processes
the full-length cDNA sequences to classify them
computational methods to infer the original transcript
isoforms. However, given the complexity of alternative against a reference annotation (e.g., GENCODE) to
identify novel genes and isoforms. The output consists
splicing, many isoforms share highly similar structures,
and the inferred transcripts are often inaccurate. of classified full-length isoforms with read counts that
are compatible with tertiary analysis software.
PacBio HiFi reads sequence full-length RNA isoforms
without the need for cDNA fragmentation and The Kinnex full-length RNA kit takes full-length cDNA
transcript assembly (Figure 1), allowing for generated by the Iso-Seq express 2.0 kit as input and
unambiguous full-length isoform detection. produces a sequencing-ready library in a two-day
workflow.
HiFi sequencing advantages for full-length
RNA sequencing
• Sequence full-length isoforms from 5’ to 3’ ends.
• Accurately characterize splice sites.
• Discover novel genes and isoforms.
• Detect variants and allele-specific isoform
expressions.
• Obtain isoform read count information.
The Iso-Seq method sequences full-length transcripts
using PacBio HiFi sequencing and has been applied to
many areas in biology and disease. It has been used in
human disease research to identify aberrant splicing
linked to rare diseases, phenotypic traits, and
neurological diseases. In cancer research, the Iso-Seq
method has been used to discover cancer-driving Figure 2. Kinnex full-length RNA sequencing. Full-length
mutations, fusion genes, and neoepitopes that could cDNA molecules are concatenated into a large-insert library,
potentially be used as cancer vaccine candidates (Li et sequenced, then processed using the PacBio software.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 3
Kinnex RNA library workflow Kinnex RNA bioinformatics workflow
The Kinnex full-length RNA workflow (Figure 3) begins The SMRT Link Read segmentation and Iso-Seq
with total RNA and produces a sequencing-ready workflow (Figure 4) processes the HiFi reads generated
library. from the Kinnex full-length RNA library to produce
classified isoforms with read counts that are
compatible with tertiary analysis tools.
Isoform clustering
FLNC reads are clustered by their sequencing similarity
to produce isoform consensus sequences. This step is
the last step of Iso-Seq analysis if no genome is
provided.
Figure 3. Kinnex full-length RNA library workflow.
Full-length cDNA molecules are synthesized with
handles (using Iso-Seq express 2.0 kit) that are
compatible with Kinnex array formation. cDNA
barcodes are added as part of the cDNA amplification
on the 5’ end. As partial arrays are removed during
nuclease digestion, Kinnex adapter ligation ensures full
arrays are enriched prior to HiFi sequencing. The
barcoded cDNAs support up to 12-plex, while the
Kinnex adapters support 4-plex at the library level.
With proper full array formation and adequate
sequencing, one SMRT® Cell on the Vega™ and Revio®
systems are expected to achieve 20–30 million and Figure 4. Kinnex full-length RNA analysis using the Read
50–60 million cDNA sequences, respectively (Table 1). segmentation and Iso-Seq workflow.
Metric Performance
Mapping
If a genome is provided, isoform consensus sequences
Sample preparation time 2 days
from the previous step are mapped and further
collapsed by their exonic structures to produce
Expected library size 11,000–18,000 bp isoforms as GFF files for visualization.
Transcript classification
Target P1 loading 60–80%
If an annotation (e.g., Gencode) is provided, isoforms
2.5–3.5 million HiFi reads (Vega) are classified against it using pigeon (the PacBio
Expected HiFi yield 6–7.5 million HiFi reads (Revio implementation of SQANTI3) to identify known and
system with SPRQ™ chemistry)
novel genes/isoforms. The Iso-Seq workflow can jointly
Expected full array % 80–90% analyze pooled sample reads to produce a unified
isoform annotation with per-sample read counts, both
20–30 million reads (Vega)
Expected read yield raw and normalized as counts per million (CPM).
50–60 million reads (Revio SPRQ)
Table 1. Target Kinnex full-length RNA library performance.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 4
Currently, SMRT Link only supports transcript While sequencing depth varies depending on the
classification for human and mouse samples. Non- experimental goal and sample, Table 3 offers some
human/mouse samples will require customized general recommendations.
annotation GTF files to be run via the command line
SMRT Link considerations Kinnex public dataset release
Below are some common considerations in running the The Kinnex RNA dataset release consists of the
Iso-Seq workflow and recommendations. Universal Human Reference RNA (UHRR), HG002 cell
line, human heart and cerebellum. Additional samples
Currently the SMRT Link Read segmentation and Iso-
from WTC-11 cell line, human brain, sorghum and
Seq workflow support human and mouse reference
mouse (to be published) are included as comparison.
genomes and annotations to produce classified
After the Read segmentation and Iso-Seq workflow,
isoforms with read counts. If working with other
each sample obtained >20,000 unique genes (Table 4).
organisms, see Table 2 for analysis recommendations.
Transcript lengths ranged from 100 bp to 11,000 bp
with minor differences that appear sample- and
Reference/annotation Analysis recommendation
species-dependent (Figure 5).
Use the Iso-Seq workflow with pre-
loaded human/mouse annotation to
Human or mouse get mapped, unique isoforms with
classifications and read count
information (FASTA, GFF, TXT).
Run Iso-Seq workflow with uploaded
reference genome to get mapped,
unique isoforms (FASTA, GFF).
Model organism with
Generate pigeon-compliant
good annotation
annotation and use the command
line for isoform classification with
read count information (TXT).
Run Iso-Seq workflow with uploaded
Non-model organism
reference genome to get mapped,
with genome
unique isoforms (FASTA, GFF).
Run Iso-Seq workflow without Figure 5. Transcript length of different Kinnex libraries shows
No genome reference genome to get unique variations that appear sample- and species-specific but are
isoforms (FASTA).
largely within the same size ranges.
Table 2. Analysis recommendations for Iso-Seq data based
on reference genome and annotation availability.
Isoform discovery and Isoform discovery of
Comprehensive transcript annotation in a
Study goal quantification of moderate-to-rare high expressed
species
transcripts transcripts
Disease vs. normal tissues with Disease cohort with
Example Plant or animal with multiple tissue types
multiple replicates >20+ samples
Target depth 10M reads per sample 5M reads per sample 5M reads per sample
3-plex cDNA for 1 Vega SMRT® Cell
6-plex cDNA for 1 Vega SMRT Cell, or
Library or
12-plex cDNA for 1 Revio SMRT Cell
6-plex cDNA for 1 Revio SMRT Cell
Read segmentation and Iso-Seq workflow with option to “pool reads and cluster together” to get a master
Analysis
isoform classification file with per-sample full-length read counts
Table 3. Example sequencing and analysis recommendations based on different study goals.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 5
Mean S-read Known Novel Known Novel
Sample Library HiFi reads S-reads
length genes genes isoforms isoforms
Kinnex-Vega 4,642,616 35,917,435 1,759 ~51,600 ~82,200
~18,600 per ~1180 per
(~5 million per per per
sample sample
sample) sample sample
UHRR Kinnex-Revio 8,247,394 63,611,648 1,753 ~60,400 ~136,800
~20,800 per ~2300 per
(6-plex (~10 million per per per
sample sample
cDNA) sample) sample sample
Kinnex-Revio 10,448,251 81,377,012 1,758 ~64,400 ~170,200
~21,760 per ~3050 per
SPRQ (~13 million per per per
sample sample
sample) sample sample
HG002 HG002 5,984,046 38,740,671 2,227 18,230 8,448 55,689 281,460
Day 0 – rep 1 6,920,750 54,110,504 1,856 19,905 3,636 58,651 212,523
Day 0 – rep 2 8,611,025 67,547,611 1,764 21,170 4,932 63,553 277,189
Day 0 – rep 3 8,124,744 63,251,235 1,864 20,744 4,822 62,349 257,665
Day 1 – rep 1 6,430,958 49,897,067 1,743 20,204 4,629 60,451 213,429
Day 2 – rep 1 7,353,759 58,217,895 1,201 21,169 6,570 64,066 165,496
WTC-11 Day 3 – rep 1 5,483,994 42,173,159 1,844 19,436 2,692 56,533 185,650
Day 3 – rep 2 6,687,580 52,317,384 1,705 21,270 4,482 63,430 241,726
Day 4 – rep 1 7,295,962 57,061,795 1,727 21,751 3,594 63,466 225,636
Day 5 – rep 1 6,645,009 51,741,094 1,751 21,754 3,195 62,217 185,807
Day 5 – rep 2 7,542,604 59,092,202 1,792 21,721 3,613 65,369 228,584
Day 5 – rep 3 6,358,300 49,466,302 1,803 21,389 3,652 59,638 187,394
Table 4. Kinnex full-length RNA dataset release for the UHRR, HG002, and additional collaborator WTC-11 cell lines. HiFi reads were
analyzed using the Read segmentation and Iso-Seq worfklow in SMRT Link v13.0 and above. All samples were Kinnex libraries and
sequenced on one SMRT Cell on Vega or Revio systems. Novel genes and isoforms are determined against Gencode v39 annotation
using pigeon. More Kinnex full-length datasets can be found at https://pacb.com/datasets.
Comparing Kinnex against non-Kinnex libraries showed number of novel isoforms detected, though most newly
that the transcript lengths did not shift when using discovered isoforms would have lower abundances.
Kinnex concatenation or different sequencing
platforms (Figure 6). Further, isoform abundances
remained consistent across sequencing platforms
(Figure 7).
Kinnex libraries showed high technical reproducibility
(Table 5), consistent with what is shown for technical
reproducibility in matching Illumina data (not shown).
Saturation curves showed that at 10 million reads,
most of the known genes and isoforms could be
detected (Figure 8, top). Alternatively, by looking at the
number of additional known isoforms detected, the
discovery rate drops below 0.1% at ~10 million reads,
i.e., for every additional 1,000 reads sequenced, one
more known isoform is detected (Figure 8, bottom). Figure 6. Transcript length distributions did not differ
Increasing sequencing depth also increases the between Kinnex (concatenated) and unconcatenated
libraries for the same UHRR sample.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 6
Figure 7. High isoform abundance correlation across
different sequencing platforms. The same UHRR library run
on Vega and the Revio (with SPRQ chemistry) system shows
high isoform count concordance.
WTC-11 day 0 WTC-11 day 5
Rep 1 Rep 2 Rep 3 Rep 1 Rep 2 Rep 3
Rep 1 1.00 0.80 0.79 Rep 1 1.00 0.80 0.80
Rep 2 0.80 1.00 0.81 Rep 2 0.80 1.00 0.79
Rep 3 0.79 0.81 1.00 Rep 3 0.80 0.79 1.00 Figure 8. Saturation and discovery curves for known isoforms of
human heart and cerebellum samples. The saturation curve (top)
for known isoforms detected at different read depths shows that
Table 5. Good technical reproducibility in Kinnex libraries. Three
80% of the known isoforms are detected at ~10M reads. The
technical replicates each from WTC-11 sample show high isoform
discovery rate for known isoforms at different read depths (bottom)
abundance correlation for both day 0 and day 5, similar to
drops below 0.1% at ~10M reads.
observed correlation values for matching Illumina technical
replicates (0.78–0.82, data not shown).
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 7
Analysis of the Kinnex RNA datasets generated on
various human samples demonstrated:
• Kinnex full-length RNA kit is compatible with all
PacBio long-read sequencing platforms
including Sequel II/IIe, Vega, and Revio
systems.
• Kinnex libraries did not alter detected transcript
sizes or abundances compared to regular Iso-
Seq libraries.
• Differences in detected transcript sizes varied
by species and sample type, but largely
remained the same in Kinnex libraries.
• High technical reproducibility across library
replicates and PacBio long-read sequencing
platforms.
• At 10 million reads, most known genes and
isoforms were detected. The number of novel
isoforms detected increases with sequencing
depth but become increasingly rare.
The Kinnex full-length RNA kit together with SMRT Link
analysis deliver high-quality, full-length that promises to
deliver extraordinary insight into disease and biology.
Conclusion
The PacBio Iso-Seq method sequences full-length
transcripts with high accuracy, enabling unambiguous
isoform characterization, allele-specific isoform
expression, differential expression analysis, and others.
The Kinnex full-length RNA kit increases throughput by
8-fold using the MAS-Seq concatenation technology.
Coupled with flexible multiplexing strategies and a
SMRT Link bioinformatics workflow, users can now
achieve comprehensive isoform sequencing in a cost-
effective manner.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Page 8
Resources & references
Resources References
Application brief – A more complete cancer Al’Khafaji, A. M., et al. (2023). High-throughput RNA
transcriptome with the Iso-Seq method – single-cell isoform sequencing using programmed cDNA
and bulk RNA sequencing. concatenation. Nature Biotechnology, 1-5.
https://doi.org/10.1038/s41587-023-01815-7
Whitepaper – Bulk and single-cell isoform sequencing
for human disease research. Li, Z., et al. (2024). An isoform-resolution transcriptomic
atlas of colorectal cancer from long-read single-cell
Application note – Bioinformatics tools for full-length
sequencing. Cell Genomics
isoform sequencing.
https://doi.org/10.1016/j.xgen.2024.100641
Kinnex website
Pardo-Palacios, F., et al. (2024). Systematic
Kinnex full-length RNA dataset: assessment of long-read RNA-seq methods for
https://pacb.com/datasets transcript identification and quantification. Nature
Biotech https://doi.org/10.1038/s41592-024-02298-3
Iso-Seq documentation: https://isoseq.how/
Pardo-Palacios, F., et al., (2024). SQANTI3: curation of
pigeon documentation:
long-read transcriptomes for accurate identification of
https://isoseq.how/classification/
known and novel isoforms. Nature Methods
https://doi.org/10.1038/s41592-024-02229-2
Stark, R., Grzelak, M., & Hadfield, J. (2019). RNA
sequencing: the teenage years. Nature Reviews
Genetics, 20(11), 631-656.
https://doi.org/10.1038/s41576-019-0150-2
Wang, B., et al. (2020). Variant phasing and haplotypic
expression from long-read sequencing in
maize. Communications Biology, 3(1), 78.
https://doi.org/10.1038/s42003-020-0805-8
Zhang, R., et al. (2022). A high-resolution single-
molecule sequencing-based Arabidopsis transcriptome
using novel methods of Iso-Seq analysis. Genome
Biology, 23(1), 149. https://doi.org/10.1186/s13059-
022-02711-0
Research use only. Not for use in diagnostic procedures. © 2024 Pacific Biosciences of California, Inc. (“PacBio”). All rights reserved.
Information in this document is subject to change without notice. PacBio assumes no responsibility for any errors or omissions in this
document. Certain notices, terms, conditions and/or use restrictions may pertain to your use of PacBio products and/or third-party
products. Refer to the applicable PacBio terms and conditions of sale and to the applicable license terms at pacb.com/license. Pacific
Biosciences, the PacBio logo, PacBio, Circulomics, Omniome, SMRT, SMRTbell, Iso-Seq, Sequel, Nanobind, SBB, Revio, Onso, Apton,
Kinnex, PureTarget, SPRQ, and Vega are trademarks of PacBio.
© 2024 PacBio. All rights reserved. Research use only. Not for use in diagnostic procedures.
102-326-591 REV03 NOV2024
Tekst er automatisk ekstrahert fra originaldokumentet. For autoritativt innhold, se originalfilen.