03_Output_Files.md

Output

TELR outputs non-referece TE insertion predictions in multiple format.

• <sample>.telr.vcf: non-reference TE insertion predictions in VCF format (1-based).
• <sample>.telr.json: non-reference TE insertion predictions in JSON format (0-based).
• <sample>.telr.expanded.json: non-reference TE insertion predictions in JSON format (0-based) with expanded information.
• <sample>.telr.bed: non-reference TE insertion predictions in BED format (0-based).
• <sample>.telr.fasta: TE insertion sequences assembled by TELR.
• <sample>.telr.contig.fasta: Local contig sequences assembled by TELR that include new TE insertions.

Note: If two or more TE families are reported for a given insertion (separated by '|' in the TE prediction output file), this could indicate a complex nested insertion event.

VCF file output by TELR

TELR generates a standard VCF file <sample>.telr.vcf in v4.1 format that has detailed information for each non-reference TE insertion.

Column	Description
chromosome	The chromosome name where the TE insertion occurred
position	Starting breakpoint position of the TE insertions.
ID	The id of the TE insertions.
Ref	The sequence of the reference is always set to N.
Alt	The TE insertion sequence.
Quality	This is currently not indicated.
Filter	This is currently always set to be PASS
Info	Provides a list of information (see below)
FORMAT	Provides information about the next tag
Sample information	Depending on the way sniffles was run: Genotype estimation:Reads supporting the reference: Reads supporting the variant.

Info field description

Sniffles report multiple information in the Info field. The entries are delimited by ;.

INFO key	Description
SVTYPE=	The type of the variant, currently this is always set to INS
END=	The position of the second breakpoint of the TE insertion
FAMILY=	TE families of the insertion, multiple families are separated by '|'
STRANDS=	Strand that TE insertion occurs
SUPPORT_TYPE=	Type of support from flank to reference alignment (single_side or both_sides)
RE=	Number of reads supporting the TE insertion
AF=	Allele frequency of the variant
TSD_LEN=	Length of the TSD sequence if available
TSD_SEQ=	TSD sequence if available

JSON file output by TELR

The VCF and JSON files are essentially equivalent, but the JSON file <sample>.telr.json can be easier to parse programmatically. For each non-reference TE insertion, the JSON file contain these keys:

Key	Description
type	The type of the TE insertions (non-reference only in current version)
ID	The unique id of the TE insertions
chrom	The chromosome name where the TE insertion occurred
start	Starting breakpoint position of the TE insertions
end	The position of the second breakpoint of the TE insertion
family	TE families of the insertion, multiple families are separated by '|'
strand	Strand that TE insertion occurs
support	Type of support from flank to reference alignment (single_side or both_sides)
tsd_length	Length of the TSD sequence if available
tsd_sequence	TSD sequence if available
te_sequence	The TE insertion sequence
genotype	Genotype of the variant
num_sv_reads	Number of reads supporting the SV allele
num_ref_reads	Number of reads supporting the reference allele
allele_frequency	Allele frequency of the variant

Expanded JSON file output by TELR

Comapred to the basic JSON file described above, the expanded JSON file <sample>.telr.expanded.json includes more QC metrics that could help with filtering TE sequences for subsquent analysis. For each non-reference TE insertion, the expanded JSON file contain these additional keys compared to the basic JSON file:

Key	Description
gap_between_flank	The size of the gap between 3' and 5' flanking sequence alignment to the reference genome (the value is negative if two alignments overlap)
te_length	The length of the new TE insertion sequence
contig_id	Unique ID for the local contig assembly
contig_length	The length of the local contig assembly
contig_te_start	Starting position of new TE insertion in the contig assembly
contig_te_end	End position of new TE insertion in the contig assembly
5p_flank_align_coord	Coordinate of 5' flanking sequence alignment to the reference genome
5p_flank_mapping_quality	Mapping quality of 5' flanking sequence alignment
5p_flank_num_residue_matches	Number of residue matches of 5' flanking sequence alignment
5p_flank_alignment_block_length	Alignment block length of 5' flanking sequence alignment
5p_flank_sequence_identity	Sequence identity of 5' flanking sequence alignment (calculated as 5p_flank_num_residue_matches/5p_flank_alignment_block_length)
3p_flank_align_coord	Coordinate of 3' flanking sequence alignment to the reference genome
3p_flank_mapping_quality	Mapping quality of 3' flanking sequence alignment
3p_flank_num_residue_matches	Number of residue matches of 3' flanking sequence alignment
3p_flank_alignment_block_length	Alignment block length of 3' flanking sequence alignment
3p_flank_sequence_identity	Sequence identity of 3' flanking sequence alignment (calculated as 3p_flank_num_residue_matches/3p_flank_alignment_block_length)

BED file output by TELR

The BED file includes minimal info about non-reference TE insertions but is easier to use in subsquent bioinformatics analysis. For each non-reference TE insertion, the BED file contain these fields:

Column	Description
Chromosome	The chromosome name where the TE insertion occurred
Start	Starting breakpoint position of the TE insertions
End	The position of the second breakpoint of the TE insertion
Family	TE families of the insertion, multiple families are separated by '|'
Score	'.'
Strand	Strand that TE insertion occurs

TE FASTA file output by TELR

For each non-reference TE insertion, TELR reports new TE insertion sequences in <sample>.telr.te.fasta.

Contig FASTA file output by TELR

For each non-reference TE insertion, TELR reports assembled contig sequences in <sample>.telr.contig.fasta.

Log file output by TELR

For each TELR run, a log file called <sample>.log is generated that records all the major steps in the program and error messages. <sample>.log: log file of TELR run.