Align mbdbs data to publicly available mito genome

[laurahspencer]

To estimate conversion efficiency

[laurahspencer]

@sr320 I presume I can simply use the file Mmag_mito.fa you added to our repo for this task?

[sr320]

Yes

[laurahspencer]

Aligned the MBDBS data to the mitochondrial genome using the script bismark_mitochondria.sh. Bismark reports for each sample and the summary report and a MultiQC report are all located in results/bismark/mito. Mitochondria % CpG methylation isn't zero, but ranges from 2.8% - 27.5% and is on average 13.9%. Mitochondrial CpG methylation percentage correlates closely with %CpG methylation across the whole genome, see figure ("included" in the figure legend means "included in analysis"):

[mgavery]

Thanks for doing this, Laura! Wow, that is def. not "zero methylation". What's really crazy for these is how high the CHG and CHH methylation is, I think CHG <2% for the nuclear genome and >20% for mito. ..more ques than answers :)

[mgavery]

sheesh.. [https://pubmed.ncbi.nlm.nih.gov/31665742/]

[sr320]

I think we need a table of CG CHG CHH methylation context for all samples, for both full genome and mitochondria to confirm sample integrity.

[laurahspencer]

Prep Sample #	Collection Sample #	pCO2	Rep	tank	Stage	DNA batch	% Alignment (genome)	% CpG meth (genome)	% CHG meth (genome)	% CHH meth (genome)	% alignment (mito)	% CpG meth (mito)	% CHG meth (mito)	% CHH meth (mito)	included in analysis?
1	CH01-06	Low	LC	3	J7	1a	31.1	73.4	1.3	7	0.11	20.6	26.5	47.8	yes
2	CH01-14	Low	LC	3	J6	3a	26.5	67.7	1.7	12.8	0.13	20.6	25.7	46.8	yes
3	CH01-22	Low	LB	1	J6	2b	30	66.2	1.2	8.2	0.12	14.6	19.9	45.3	yes
4	CH01-38	Low	LB	1	J6	1b	30.4	68.1	1.6	12.2	0.13	24.8	31.2	48.3	yes
7	CH03-33	Low	LB	1	J6	3b	30.8	72.5	1.1	5.6	0.09	11.6	16.8	44.7	yes
12	CH07-06	Low	LC	3	J7	2a	30.7	64.1	1.3	9	0.13	13	19.7	45.4	yes
10	CH05-21	High	HB	2	J7	3a	31.1	64.1	1.3	12.5	0.12	12.2	18.7	47.1	yes
11	CH05-24	High	HC	4	J6	1a	31.5	71.6	1.6	13.9	0.13	27.5	31.4	50.3	yes
15	CH09-02	High	HC	4	J6	3b	29.7	68.4	1.8	15.6	0.16	23.1	26.4	47.6	yes
17	CH09-28	High	HA	6	J6	4a	31.7	67.6	1.7	7.8	0.13	14.8	22.9	46.3	yes
18	CH10-01	High	HB	2	J6	4a	35.3	70.8	1.4	7.7	0.16	19.2	26	47.6	yes
20	CH10-11	High	HA	6	J6	3b	31.9	67.7	1.5	12.3	0.16	21.9	26.6	46.1	yes
5	CH03-04	Low	LB	1	J7	1b	23.5	11.4	2.5	35.4	0.16	5	9.1	40.8	no
6	CH03-15	Low	LB	1	J6	2a	19.1	19.4	1.5	18.4	0.13	4.4	8.7	37.5	no
13	CH07-11	Low	LC	3	J6	3a	18.1	7.8	1.6	12.7	0.12	3.5	7	33.1	no
14	CH07-24	Low	LB	1	J7	2a	18.1	7	1.6	13.7	0.15	5.1	9.7	37.4	no
8	CH05-01	High	HB	2	J7	1a	28.2	51.8	2.1	26.6	0.14	16.2	25	48.2	no
9	CH05-06	High	HA	6	J7	3a	16	7	1.4	16	0.12	2.8	5.5	30.9	no
16	CH09-13	High	HA	6	J7	2b	12	45.8	1.9	19.7	0.05	10.4	13.6	41	no
19	CH10-08	High	HA	6	J6	2b	27.9	31.2	1.8	26.8	0.14	6.7	13	43.9	no

[laurahspencer]

Also uploaded the table to the repo here: results/alignment-meth-summary.csv

[laurahspencer]

Simpler table sorted by genome CpG % methylation rate:

Sample	pCO2	Genome Alignment %	Genome % CpG	Genome % CHG	Genome % CHH	Mito Alignment %	Mito CpG %	Mito CHG %	Mito CHH %	included?
1	L	31.1	73.4	1.3	7	0.11	20.6	26.5	47.8	yes
7	L	30.8	72.5	1.1	5.6	0.09	11.6	16.8	44.7	yes
11	H	31.5	71.6	1.6	13.9	0.13	27.5	31.4	50.3	yes
18	H	35.3	70.8	1.4	7.7	0.16	19.2	26	47.6	yes
15	H	29.7	68.4	1.8	15.6	0.16	23.1	26.4	47.6	yes
4	L	30.4	68.1	1.6	12.2	0.13	24.8	31.2	48.3	yes
20	H	31.9	67.7	1.5	12.3	0.16	21.9	26.6	46.1	yes
2	L	26.5	67.7	1.7	12.8	0.13	20.6	25.7	46.8	yes
17	H	31.7	67.6	1.7	7.8	0.13	14.8	22.9	46.3	yes
3	L	30	66.2	1.2	8.2	0.12	14.6	19.9	45.3	yes
10	H	31.1	64.1	1.3	12.5	0.12	12.2	18.7	47.1	yes
12	L	30.7	64.1	1.3	9	0.13	13	19.7	45.4	yes
8	H	28.2	51.8	2.1	26.6	0.14	16.2	25	48.2	no
16	H	12	45.8	1.9	19.7	0.05	10.4	13.6	41	no
19	H	27.9	31.2	1.8	26.8	0.14	6.7	13	43.9	no
6	L	19.1	19.4	1.5	18.4	0.13	4.4	8.7	37.5	no
5	L	23.5	11.4	2.5	35.4	0.16	5	9.1	40.8	no
13	L	18.1	7.8	1.6	12.7	0.12	3.5	7	33.1	no
14	L	18.1	7	1.6	13.7	0.15	5.1	9.7	37.4	no
9	H	16	7	1.4	16	0.12	2.8	5.5	30.9	no

[laurahspencer]

Mitochondrial alignment is not helpful for conversion efficiency estimation, but is helpful to look at consistency across samples.