File formats for Sequence data analysis

Abel Vertesy

File formats in sequence analysis

pipeline {width=300px}

FastA format to store sequences (and align them)

>SequenceID
SEQUENCE (ATGCAGTATAG)
Used for proteins too
>gi|5524211|gb|AAD44166.1| cytochrome b [Elephas maximus maximus]
LCLYTHIGRNIYYGSYLYSETWNTGIMLLLITMATAFMGYVLPWGQMSFWGATVITNLFSAIPYIGTNLV
EWIWGGFSVDKATLNRFFAFHFILPFTMVALAGVHLTFLHETGSNNPLGLTSDSDKIPFHPYYTIKDFLG
LLILILLLLLLALLSPDMLGDPDNHMPADPLNTPLHIKPEWYFLFAYAILRSVPNKLGGVLALFLSIVIL
GLMPFLHTSKHRSMMLRPLSQALFWTLTMDLLTLTWIGSQPVEYPYTIIGQMASILYFSIILAFLPIAGX
IENY
  • The reference Genome / Transcriptome will also be stored as FASTA!

FastQ (after sequencing)

Name: Fasta with Quality-score from the sequencer

FastQ is the most raw form of scRNASeq data you will encounter.

For paired-end sequencing, you will have 2 .fastq files with matched names, and matched reads, line-by line

FastQ files have the format:

>ReadID
READ SEQUENCE
+
SEQUENCING QUALITY SCORES

SAM (after mapping)

Name: Sequence Alignment Map

A header, which typically includes information on the sample preparation, sequencing and mapping;

A body: a tab-separated row for each individual alignment of each read.

Header
listing all chromosomes or transcripts (depending on your reference)

Body 
1 read per line + a lot of additional fields
Example
@SQ    SN:chr1    LN:50
@SQ etc...


1:497:R:-272+13M17D24M  113  1  497  37  37M  15  100338662  0  CGGGTCTGACCTGAGGAGAACTGTGCTCCGCCTTCAG  0;==-==9;>>>>>=>>>>>>>>>>>=>>>>>>>>>>  XT:A:U  NM:i:0  SM:i:37  AM:i:0  X0:i:1  X1:i:0  XM:i:0  XO:i:0  XG:i:0  MD:Z:37

1:497:R:-272+13M17D24M  113  1  497  37  37M  15  100338662  0  CGGGTCTGACCTGAGGAGAACTGTGCTCCGCCTTCAG  0;==-==9;>>>>>=>>>>>>>>>>>=>>>>>>>>>>  XT:A:U  NM:i:0  SM:i:37  AM:i:0  X0:i:1  X1:i:0  XM:i:0  XO:i:0  XG:i:0  MD:Z:37

Interpreting the .sam file

Field name description Example-data
QNAME read name 1:497:R:-272+13M17D24M
FLAG alignment flag 113
RNAME alignment chromosome 1
POS alignment start position 497
MAPQ overall mapping quality 37
CIGAR alignment CIGAR string 37M
MRNM/RNEXT name of next align. … 15
MPOS/PNEXT pos. of next alignm. … 100338662
ISIZE/TLEN observed Template LENgth 0
SEQ sequence CGGGTCTGACCTGAGGAGAACTGTGCTCCGCCTTCAG
QUAL quality per base 0;==-==9;>>>>>=>>>>>>>>>>>=>>>>>>>>>>
TAGs further tags with alignment info XT:A:U NM:i:0 SM:i:37 AM:i:0 X0:i:1 X1:i:0 XM:i:0 XO:i:0 XG:i:0 MD:Z:37

Important fields:

  • Read length → What sequencer was used?
  • Trimming bases → QUAL
  • Filtering out and reads → MAPQ, TAGs
    • Not align robustly aligned

BAM

The machine-readable version of SAM file and it is highly compressed.

BAM/SAM files can be converted to the other format using ‘samtools’:

BAM files can be converted back to FastQ using bedtools.

What can you do with .sam file?

  1. You can filter out reads that are not aligned with a high quality

    1. Cannot be mapped
    2. Maps to mult
      1. The shorter the read the more problematic this is

  2. Based on .sam files, you can count your reads per gene

    → See: Construction of expression matrix

  3. Create pileup file

Use pileup files to see genetic variation and sequencing errors per position

Pileup

Other formats: GFF, GTF

Genome annotation files, used to create a genome index or reference transcriptome from the reference genome.

GTF files contain annotations of genes, transcripts, and exons. They must contain: