Single-Cell RNA Sequencing using the Iso-Seq Method

Elizabeth Tseng, Principal Scientist, PacBio

SINGLE-CELL ISO-SEQ METHOD: WHY?

PACBIO HIFI READS ARE ACCURATE FOR SINGLE-CELL

COMPATIBLE WITH FULL-LENGTH SINGLE-CELL PLATFORMS

SINGLE-CELL ISO-SEQ WORKFLOW

• Compatible with any full-length single-cell platform
• Detailed workflow guidance
• Standard Iso-Seq library preparation & sequencing workflow
• Can generate matching short read data from same sample
• ~3 million HiFi reads per SMRT Cell 8M
• Flexible barcoding & multiplexing based on desired number of reads per cell

Single-Cell Iso-Seq Research Highlights

SINGLE-CELL ISO-SEQ METHOD ON DIFFERENT PLATFORMS

COMBINING SHORT- AND LONG-READ SINGLE CELL DATA

• Short read data identifies cell clusters
• Long read data identifies full-length isoforms
• Matching (UMI, BC) links isoforms to cell types

Click here for more information

CHIMP-HUMAN TSS DIFFERENCE IN SINGLE-CELL ORGANOIDS

Click this Youtube video if you want to know more

AGED MICE SHOW MORE DIVERSE V(D)J RECOMBINATION

CONCATENATION OF SINGLE-CELL TRANSCRIPTS

Click here for the article

Human influenza infections are initiated by just a few virions

Click here and here for the articles

1. We take many cells, and infect them with many virions
2. We usually average over this entire process to study infection
3. We infect A549 cells at low multiplicity of infection (MOI)
4. This yields a large cell-gene matrix that we cananalyze computationally

• Extreme variation across single cells

1. I said that we used a stock of “wildtype” virus. Actually, all viral stocks contain mutants. Maybe mutations in the virions explain variation across cells…
   • Single-cell transcriptomics counts mRNAs, it doesn’t tell us if they have mutations

1. We start with the same process as for single-cell transcriptomics. But we also perform full-length PacBio sequencing on viral genes

• Cells infected by wildtype virions often produce lots of viral mRNA

• Although some wildtype virions produce less viral mRNA than others

• Virions with mutations sometimes produce little viral mRNA

• Virions with defects sometimes produce IFN: fails to express NS

• Virions with defects sometimes produce IFN: point mutation in NS

• Virions with defects sometimes produce IFN: internal deletion in PB1

• Virions with defects sometimes produce IFN: point mutation in PB1

• But even virions lacking NS do not always induce IFN

• And sometimes wildtype virions induce IFN

• 150 cells infected, only 49 by wildtype virions

• Infections by wildtype virions are less heterogeneous than ones by mutant ones

• Infections by mutant virions induce more IFN

SINGLE-CELL VIRAL SEQUENCING

• Single-cell sequencing of H1N1 influenza
• Full-length viral transcripts reveal cell-to-cell variation on mutational landscape
• 10hr post-infection, only 49 of 150 infected cells remain wild type
• Mutations linked to differences in viral load and the innate immune response

Single-Cell Iso-Seq Bioinformatics

SINGLE-CELL ISO-SEQ BIOINFORMATICS WORKFLOW

1. HiFi reads

1. Remove cDNA primers

1. Extract UMI and BC

1. Cluster by (UMI,BC) (Click here for the GitHub page)

1. Classify Transcripts

2. Remove Artifacts (Click here for the GitHub page, and here for more about SQUANTI3)

TRANSCRIPT CLASSIFICATION BY SQANTI

SQANTI3: BEFORE AND AFTER FILTERING

• After SQANTI3 filtering, percentage of FSMs increase
• Most filtered transcripts are genic, genomic, or intergenic

SINGLE-CELL ISO-SEQ BIOINFORMATICS WORKFLOW

Click here to get more information about how this data was generated

The pdf to this documentation can be found here

Special Iso-Seq topic: Single Cell Iso-Seq (hands on)

The Iso Seq Bioinformatics Tutorial we are using is here. See ‘4. Single Cell’ and answer the corresponding practice questions!