Day 1 – RNA-seq library prep and Trimming

Presented by Beant Kapoor

08-14-2023

Why is RNA seq done?

• To calculate gene expression
• Protein coding transcripts or messenger RNA (mRNA)
• mRNA - 1-5% of total RNA
• rRNA - 80-90%

RNA-seq library prep

1. Depletion or Enrichment
2. Conversion of RNA to cDNA
3. Add sequencing adapters

TruSeq and SMART-seq library preps

RNA enrichment

mRNA

PolyA capture

rRNA depletion

TruSeq RNA sample prep

Enriched RNA

Fragment and prime

Reverse transcription

Second strand synthesis

TruSeq RNA sample prep

SMART-seq sample prep

Reverse transcription

Template switch

cDNA PCR

DNA library preparation

Trimming

• Removing low quality data or portions of sequence data might interfere with our analyses
• Why trim?
• Erroneous sequencing calls (low Q scores)
• Read contains synthetic oligos from library preparation

Example Fastq file

How is a Q-score calculated?

Q = -10log₁₀P

Let’s say P = 0.01

1 in 100 probability, base is miscalled

Q = -10log₁₀(0.01)

Q = -10 x -2

Q = 20

How is a Q-score calculated?

Q = -10log₁₀P

Let’s say P = 0.001

1 in 1,000 probability, base is miscalled

Q = -10log₁₀(0.001)

Q = -10 x -3

Q = 30

​

How is a Q-score calculated?

illumina.com

How is a Q-score calculated?

illumina.com

RTA2

RTA3

How reliable is a Q-score?

illumina.com

Summary - when to trim?

• When you know your data has adapter contamination
• If you’re performing genome sequencing or SNP-calling
• If sequence data, on average, have low-quality scores
• If you have many reads to spare