ctg_hex_pos and ctg_hex_dist is buggy when the contig has indels

[zyxue]

ctg_hex_pos and ctg_hex_dist is buggy when the contig has indels

[zyxue]

Add an example

index	seqname	strand	clv	aclv	gene_name	gene_id	signed_dist_to_aclv	evidence_type	contig_ids	max_contig_len	max_contig_mapq	any_contig_is_hardclipped	num_suffix_contigs	num_bridge_contigs	num_link_contigs	num_blank_contigs	num_total_contigs	num_suffix_reads	max_suffix_contig_tail_len	num_bridge_reads	max_bridge_read_tail_len	num_link_reads	ctg_hex	ctg_hex_id	ctg_hex_pos	ctg_hex_dist	ref_hex	ref_hex_id	ref_hex_pos	ref_hex_dist
37378	chr1	+	204489245	204501779	MDM4	ENSG00000198625	-12534	bridge	A0.R100625	3334	40	True	0	1	0	0	1	0	0	6	4	0	ACTAAA	6	204489235	10	ACTAAA	6	204489238	7
37379	chr1	+	204489253	204501779	MDM4	ENSG00000198625	-12526	bridge	A0.R100625	3334	40	True	0	1	0	0	1	0	0	35	6	0	AATACA	9	204489241	12	AATACA	9	204489244	9
46052	chr7	-	109555254	109599274	EIF3IP1	ENSG00000237064	-44020	bridge	A0.S61128	68	1	True	0	1	0	0	1	0	0	1	6	0	TTTAAA	3	109555268	14	AATAAA	16	109555282	28

ctg_hex_dist and ref_hex_dist don't match because of 3bp insertion in A0.R100625

[zyxue]

Actually, ctg_hex_dist could be correct, it's just the ctg_hex_pos is not consistent with genomics coordinates and should be dropped.

[zyxue]

Insight: It's that ctg_hex_pos isn't interpretable wst. genome coordinate. But ctg_hex_dist is probably valid.

[zyxue]

	seqname	strand	clv	aclv	gene_name	gene_id	signed_dist_to_aclv	evidence_type	contig_ids	max_contig_len	max_contig_mapq	any_contig_is_hardclipped	num_suffix_contigs	num_bridge_contigs	num_link_contigs	num_blank_contigs	num_total_contigs	num_suffix_reads	max_suffix_contig_tail_len	num_bridge_reads	max_bridge_read_tail_len	num_link_reads	ctg_hex	ctg_hex_id	ctg_hex_pos	ctg_hex_dist	ref_hex	ref_hex_id	ref_hex_pos	ref_hex_dist
39789	chr1	+	204489245	204501779	MDM4	ENSG00000198625	-12534	bridge	A0.R100625	3334	40	True	0	1	0	0	1	0	0	6	4	0	ACTAAA	6	204489235	10	ACTAAA	6	204489238	7
39790	chr1	+	204489253	204501779	MDM4	ENSG00000198625	-12526	bridge	A0.R100625	3334	40	True	0	1	0	0	1	0	0	35	6	0	AATACA	9	204489241	12	AATACA	9	204489244	9
46413	chr7	-	109555254	109599274	EIF3IP1	ENSG00000237064	-44020	bridge	A0.S61128	68	1	True	0	1	0	0	1	0	0	1	6	0	AATATA	10	109555301	47	AATAAA	16	109555282	28

Happy to see the PAS hexamers finally match between contig and reference in the third case.

Conclusion:
ctg_hex_pos is not interpretable in terms of reference genome, but decided to leave it here for reference as a sanity check/indication of insertion, so is the difference between ctg_hex_dist and ref_hex_dist.

ctg_hex_pos may become useful when the ref_hex is not found.