🦟

FastQC

2 min read

rna-seq

Workshop post

  • quick overview of any likely sequencing problems
  • summary graphs and tables to quickly access your data
  • export of results as an HTML-based report

Manual

FASTQ file format

Consists of four lines. Has a header starting with @.

LineDescription
1always begins with @ and then information about the read
2the actual DNA sequence
3always begins with + and sometimes the same info in line 1
4has a string of characters which represent quality scores; must have same number of characters as line 2

Different quality encoding scales exist (differing by offset in the ASCII table), but note the most commonly used is fastqsanger by Illumina since mid-2011.

Quality encoding: !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHI
                  |         |         |         |         |
   Quality score: 0........10........20........30........40

Each quality score represents the probability that the corresponding nucleotide call is incorrect where is the probability that a base call is erroneous.

Phred quality score of 10 corresponds to 90% base call accuracy, 20: 99%, 30: 99.9%, 40: 99.99%.

HTML report

  • Per base sequence quality
  • Per sequence quality scores
  • Per base sequence content
    • distribution of A, T, C, G across reads, per base location
    • always FAIL/WARNING, because first 10-12 bases result from the “random” hexamer priming
  • Per sequence GC content
    • in general, a central peak corresponds to the expected % GC for the organism
    • distribution should be normal, unless overrepresented sequences (sharp peaks on a normal distribution; causes saturation of other genes’ expression) or contamination with another organism (broad peak)

NOTE

Guanine-Cytosine base pairs form three hydrogen bonds (compared to two in AT), which is more stable. There is a positive correlation between GC content and gene expression level.

  • Sequence duplication levels
    • number of duplicated sentences in the library
    • number of PCR cycles and amount of input are controlled only during library prep, can just detect a low complexity library during RNA-seq if there are too many duplicates
  • Overrepresented sequences
    • at least 20 bp sequences that occur in more than 0.1% of the total number of reads
    • detect contamination of vector or adapter sequences: if GC % was off, this table can help identify the source

Graph View

Backlinks