NGS Quality Control and preprocessing

Universal Raw Data Processing Concepts

Estimating base quality: Common representation of base quality | fasta, qual, and fastq files

fastq

Paired end sequencing: Illumina

• Insert size of 200-800 bases

​

• If 200-250 bp then small over-lapping contigs can be created before assembly.

​

• Pair end sequences with known insert sizes can help bridge difficult to resolve section of DNA (repeats)

​

• Pair end sequences can be used informatically as a ‘double’ check.

​

• Either 2xFastq files created or these can be spliced together to give reverse and forward of same fragment consecutively on the sequence

First question a bioinformatician will always ask you:

​

“Single or paired?”

Trimming of introduced Sequences: �Multiplexing Identifier Adaptors (MID tags)

Commonly referred to as

Indexes or Barcodes or MIDs or Tags

Usually done by sequencing centre, but not always.

We will look at different ways to import into Qiime later!

Trimming of Introduced sequences:�Biggest risk: Sequencing barcodes or adapters

400bp

250bp

250bp

Next Generation Sequencing Platforms:�Platform Heterogeneity

Illumina

Nanopore

Pacific Biosystems

Platform specific errors and adaptor sequences

Assessing Sequence data quality

FastQC

Qiime

(We’ll look at both of these in the practical sessions)

Quality filtering

? Trim first base <30

Base filtering

Quality filtering

Trim if average of 5bp <30

Sliding window Trimming

Overlapping reads

250bp

250bp

450bp Amplicon region

trim

Quality Controlling your data in Qiime2 (Dada2)