How to read this page

Each rejections status entry in this page has the following properties:

• Level: Item category (projection, transcript, or gene) to which the rejection reason applies. Stated in the first column in rejected_items.tsv.
• Loss status: Loss status corresponding to the rejection reason. Certain rejection reason do not affect the items’ loss status; for example, projections discarded past the alignment step and transcript discarded past the loss_summary step retain their defined statuses after rejection.
• Comment text: A short comment stated in the fourth column of rejected_items.tsv corresponding to the rejection reason.
• Rejection step: Pipeline step at which items can be rejected due to the stated reason.

Rejection reasons

ALL_ORTHS_DISCARDED

Level: PROJECTION
Loss status: Highest at the time of rejection
Comment text: No projections reached the orthology step
Rejection step: orthology
Transcripts for which all projections were discarded at the orthology classification level, e.g due to weak edge support in the orthology graph.

ALL_PARALOGS_REDUNDANT

Level: TRANSCRIPT
Loss status: M
Comment text: All paralogous projections outcompeted by orthologous predictions of other items
Rejection step: loss_summary
Transcripts that do not have any orthologous projections and whose paralogous projections were outcompeted by orthologous items of other transcripts.

CHAIN_LIMIT_EXCEEDED

Level: PROJECTION
Loss status: N
Comment text: Number of homologous chains exceeds the set limit (X)
Rejection step: preprocessing
Extra chains exceeding the maximum number of chains per transcript (X). For each transcript, chains valid for projection annotation are sorted by chain ID, with items coming after the Xth chain discarded without any further annotation. The X parameter can be controlled with the --max_chains_per_transcript option (100 chains per transcript be default).

CHIMERIC

Level: PROJECTION
Loss status: N
Comment text: Potential chimeric projection
Rejection step: alignment
Projections overlapping two or more non-overlapping projections of the same reference transcript.

CHROM_UNALIGNED

Level: TRANSCRIPT
Loss status: M (previously N)
Comment text: No chain corresponds to reference chromosome
Rejection step: feature_extraction
Transcripts for which the whole reference chromosome (contig, scaffold, etc.) is missing from the genome alignment chains. Loss status was set to N in the earlier TOGA2 versions. Appears only if legacy feature extraction procedure is enabled.

EXCEEDS_MEMORY:X

Level: PROJECTION
Loss status: L
Comment text: No exons were aligned
Rejection step: preprocessing
Rejection due to memory limit per exon/exon group exceeded repeatedly for this projection. Number X corresponds to the number of exons/exon groups for this projection which were discarded from the alignment step due to this reason.

EXCEEDS_MEMORY+GAP:X

Level: PROJECTION
Loss status: M
Comment text: No exons were aligned
Rejection step: preprocessing
Same as EXCEEDS_MEMORY, but the query locus also overlaps an assembly gap.

EXCEEDS_SPACE:X

Level: PROJECTION
Loss status: L. Comment text: No exons were aligned
Rejection step: preprocessing
Rejection due to the orthologous space in the query being too long. Number X corresponds to the number of exons/exon groups for this projection which were discarded from the alignment step due to this reason.

EXCEEDS_SPACE+GAP:X

Level: PROJECTION
Loss status: M
Comment text: No exons were aligned
Rejection step: preprocessing
Same as EXCEEDS_SPACE, but the query locus also overlaps an assembly gap.

GENE_TREE_REJECTION

Level: TRANSCRIPT
Loss status: Highest at the time of rejection.
Comment text: Rejected after the gene resolution step
Rejection step: orthology
Transcripts left without any orthologs in the query after gene tree-based orthology resolution step.

HEAVY

Level: PROJECTION
Loss status: N
Comment text: Maximum memory limit exceeded
Rejection step: preprocessing
Projections for which estimated alignment memory requirements exceeded the user-defined status. Memory limit corresponds to the highest memory bin defined.

ILLEGAL_NAME

Level: TRANSCRIPT
Loss status: N
Comment text: Illegal character used in transcript name
Rejection step: setup
Reference transcripts containing illegal characters in their names.

INSUFFICIENT_CHAIN_SCORE

Level: PROJECTION
Loss status: L
Comment text: Chain score below set threshold (X)
Rejection step: classification
Projections whose chain score is lower than the minimal chain score accepted for orthology classification (--min_orthologous_chain_score). This rejection reason occurs if min_orthologous_chain_score > min_chain_score.

INSUFFICIENT_SEARCH_SPACE:X

Level: PROJECTION
Loss status: L
Comment text: No exons were aligned
Rejection step: preprocessing
Rejection due to the query locus being too small to accommodate for the reference exon alignment. Number X corresponds to the number of exons/exon groups for this projection which were discarded from the alignment step due to this reason.

INSUFFICIENT_SEARCH_SPACE+GAP:X

Level: PROJECTION
Loss status: M
Comment text: No exons were aligned
Rejection step: preprocessing
Same as INSUFFICIENT_SEARCH_SPACE, but the query locus also overlaps an assembly gap.

MULTIPLE_ORTHOLOGY

Level: TRANSCRIPT
Loss status: M (previously N) Comment text: Multiple orthologs detected
Rejection step: preprocessing
Transcripts for which more than one orthologous chain was detected. Loss status was set to N in the earlier TOGA2 versions (before v2.0.7). Appears only if --one2ones_only is set.

NON_CODING

Level: TRANSCRIPT
Loss status: N Comment text: Does not have a coding sequence
Rejection step: setup
Reference transcripts with no annotated coding sequence according to input BED12 file.

NO_CHAINS

Level: TRANSCRIPT
Loss status: M (previously N)
Comment text: No covering chains detected
Rejection step: preprocessing
Projections with absolutely no classifiable (orthologous/paralogous/spanning/processed pseudogene) chains which somehow reached the preprocessing step. Before v2.0.7, were classified as Undefined.

NO_PROJ

Level: TRANSCRIPT
Loss status: M
Comment text: No classifiable projections found
Rejection step: classification
Transcripts with no projections eligible for classification (e.g., due to insufficient chain scores).

[!NOTE] Between versions 2.0.2 and 2.0.6 included, transcripts with no projections were missing from the rejection file; as a result, respective genes would get loss status N instead of M in loss_summary.tsv.

OUT_OF_FRAME

Level: TRANSCRIPT
Loss status: N
Comment text: Transcript is out of frame
Rejection step: setup
Reference transcripts with coding sequence length not divisible by three, indicating reading frame shift or an incomplete reading frame.

REDUNDANT

Level: PROJECTION
Loss status: N
Comment text: Redundant projection to the given locus
Rejection step: alignment
Projections overlapping other projections of the same reference transcript and exceeding maximum memory ceiling for the alignment jobs.

REDUNDANT_PARALOG

Level: PROJECTION
Loss status: Defined at the alignment step
Comment text: Redundant paralog overlapping orthologous projections
Rejection step: gene_inference
Paralogous projections overlapping orthologous predictions in the query.

REDUNDANT_PPGENE

Level: PROJECTION
Loss status: Defined at the alignment step Comment text: Processed pseudogene overlapping ortholog or paralog
Rejection step: gene_inference
Processed pseudogene projections overlapping orthologous or paralogous predictions in the query.

REJECTED_CONTIG

Level: TRANSCRIPT
Loss status: N
Comment text: Located outside of user-preferred contigs
Rejection step: setup
Reference transcripts located on contigs (scaffolds, chromosomes, etc.) deprecated by the user.

SECOND_BEST

Level: PROJECTION
Loss status: Defined at the alignment step
Comment text: Insufficiently covered exons in second-best projection
Rejection step: gene_inference
Orthologous projection with orthology probability lower than the highest for a given reference transcript that overlaps top orthologous prediction for this or other transcript.

SPANNING

Level: PROJECTION
Loss status: L or M
Comment text: Spanning chain
Rejection step: preprocessing
Projection for which no coding bases are aligned in the genome alignment chain. Loss status depends on whether the expected locus overlaps with an assembly gap in the query.

TRANSCRIPT_UNALIGNED

Level: TRANSCRIPT
Loss status: M (previously N)
Comment text: No chain corresponds to the transcript
Rejection step: feature_extraction
Transcripts for which no overlapping chains were found. Loss status was set to N in the earlier TOGA2 versions. Appears only if legacy feature extraction procedure is enabled.

WEAK_EDGE

Level: PROJECTION
Loss status: Defined at the alignment step
Comment text: Weak orthology graph edge
Rejection step: orthology
Projection discarded during orthology graph reduction due to insufficient orthology probability.

ZERO_ORTHOLOGY

Level: TRANSCRIPT
Loss status: M (previously N)
Comment text: No orthologous chains detected
Rejection step: preprocessing
Transcripts with no orthologous chains detected. Relevant only if --orthologs_only option is set. Before v2.0.7, were classified as Undefined.