add indicator for stop codon in mini-map

[kwongj]

Example abricate output:

2012-17035.fna	gnl|Prokka|2012-17035_4	51134	51309	aadA1	1-184/972	===...../......	8	18.11	94.565
2012-17035.fna	gnl|Prokka|2012-17035_2	27659	27769	aadA1	679-789/789	............===	0	14.07	99.099
2012-17035.fna	gnl|Prokka|2012-17035_4	52211	53002	aadA2	1-792/792	===============	0	100.00	99.874
2012-17035.fna	gnl|Prokka|2012-17035_4	61539	62354	aph(3')-Ia	1-816/816	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_2	37261	38142	blaKPC-2	1-882/882	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_2	26775	27614	blaOXA-9	1-840/840	===============	0	100.00	99.881
2012-17035.fna	gnl|Prokka|2012-17035_1	2712753	2713613	blaSHV-11	1-861/861	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_3	6395	7237	blaSHV-12	1-861/861	========/======	18	97.91	97.909
2012-17035.fna	gnl|Prokka|2012-17035_2	25215	26075	blaTEM-1A	1-861/861	===============	0	100.00	99.884
2012-17035.fna	gnl|Prokka|2012-17035_4	44636	45295	catA1	1-660/660	===============	0	100.00	99.848
2012-17035.fna	gnl|Prokka|2012-17035_4	51306	51803	dfrA12	1-498/498	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_1	4674590	4675009	fosA	1-420/420	===============	0	100.00	98.571
2012-17035.fna	gnl|Prokka|2012-17035_4	59641	60562	mph(A)	1-921/921	========/======	1	100.00	99.675
2012-17035.fna	gnl|Prokka|2012-17035_4	59657	60562	mph(A)	1-906/906	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_1	1200950	1202125	oqxA	1-1176/1176	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_1	1197774	1200926	oqxB	1-3153/3153	===============	0	100.00	100.000
2012-17035.fna	gnl|Prokka|2012-17035_4	53420	54346	sul1	1-927/927	===============	0	100.00	100.000

However, blaOXA-9 has a TAG stop codon in the middle, probably rendering the gene non-functional. It would be useful feature to see this in the mini-map.

(Ariba currently detects this and outputs it in the table.)

[tseemann]

To be able to do this, I need to know which database sequences were actually CDS or ORFs. Most of them should be, but we don't always know what frame they are in unfortunately.

I need to think about how best to do this. ARIBA does it by making the user say which are CDS and which are not. I guess I need to detect this at database creation time.