mDNACN

Association of mitochondrial DNA CN with Alzheimer's disease

Ricardos AMPAD files

/sc/orga/projects/LOAD/Data/AMPAD

Location of AMPAD bim files.

/sc/orga/projects/AMPADWGS
/sc/orga/projects/AMPADWGS/RawData
/sc/orga/projects/AMPADWGS/RawDataSinai
/sc/orga/projects/AMPADWGS/RawDataMayo

/sc/orga/projects/LOAD/shea/Projects/mDNACN

ROS/MAP

cogdx

Value	Coding	n
1	NCI: No cognitive impairment (No impaired domains)	374
2	MCI: Mild cognitive impairment (One impaired domain) and NO other cause of CI	273
3	MCI: Mild cognitive impairment (One impaired domain) AND another cause of CI	20
4	AD: Alzheimer’s dementia and NO other cause of CI (NINCDS PROB AD)	433
5	AD: Alzheimer’s dementia AND another cause of CI (NINCDS POSS AD)	57
6	Other dementia: Other primary cause of dementia	21

msex

Value	n
F	779
M	399

race

Value	Coding	n
1	White	1177
2	Black	1

apoe_genotype

Value	n
22	7
23	146
24	22
33	709
34	263
44	18
NA	13

MSBB

NP.1 Neuropathology Category as measured by CERAD

Value	Coding	n
1	Normal	91
2	Definite AD	161
3	probable AD	41
4	possible AD	40

CDR Clinical Dementia Rating

Value	Coding	n
0	no dementia	39
0.5	Questionable dementia	43
1	Mild dementia	36
2	Moderate Dementia	41
3	Severe dementia	74
4	Profound dementia	45
5	Terminal Dementia	55

bbscore - Braak Score

Value	Coding	n
0		13
1	I - transentorhinal	27
2	II - transentorhinal	43
3	III - limbic	49
4	IV - limbic	32
5	V - neocortical	37
6	VI - neocortical	100
NA		22

NIA-Reagan (Kurt)

CERAD / Braak	0	1	2	3	4	5	6
Normal	Normal	Other	Other	Other	Other	Other	Other
Possible	Low	Low	Low	Low	Low	Low	Low
Probable	Low	Low	Low	Intermediate	Intermediate	Intermediate	Intermediate
Definite	Low	Low	Low	Intermediate	Intermediate	High	High

NIA-Reagan

CERAD / Braak	1	2	3	4	5	6
Normal	No AD	No AD	-	-	-	-
Possible	Low	Low	-	-	-	-
Probable	-	-	Intermediate	Intermediate	-	-
Definite	-	-	-	-	High	High

CERAD / Braak	0	1	2	3	4	5	6	NA
Normal	12	21	30	20	2	NA	NA	6
Possible	1	6	4	16	10	2	1	NA
Probable	NA	NA	4	12	6	8	8	3
Definite	NA	NA	5	11	14	27	91	13

Value	Coding	CERD	Braak	n
0	No AD	Normal	I/II
1	Low	Possible AD	I/II	27
2	Intermediate	Probable AD	III/IV	43
3	High	Definite AD	V/V1	49

NIA-AA

Thal	CERAD	Braak:	None or I/II	III/IV	V/VI
0	None		Other§	Other§	Other§
1/2	None - Sparse		Low	Low	Low¶
	Modearte - Frequent		Low†	Intermediate	Intermediate¶
3	Any		Low†	Intermediate	Intermediate¶
4/5	None - Sparse		Low†	Intermediate	Intermediate¶
	Modearte - Frequent		Low†	Intermediate	High

^† § ¶ SEX

Value	n
F	216
M	117

apoe

Value	n
22	2
23	19
24	1
33	96
34	53
44	3
NA	159

rs429358	rs7412	Name
19:45411941:T:C	19:45412079:C:T	NA
C	T	ε1
T	T	ε2
T	C	ε3
C	C	ε4

RACE

Value	Coding	n
A	Asian	1
B	Black	31
H	Hispanic	23
U	Uknown	1
W	White	277

dx

Value	n
AD	156
CONTROL	55
OTHER	122

Mayo

diagnosis

Value	n
AD	92
Control	100
Pathologic Aging	74
PSP	83

Sex

Value	n
F	182
M	167

ApoE

Value	n
22	1
23	36
24	2
33	176
34	69
44	9
NA	56

WGS_Source_Tissue_Type

Value	n
Cerebellar Cortex	8
Temporal Cortex	341

fastMitoCalc

https://lgsun.irp.nia.nih.gov/hsgu/software/mitoAnalyzer/index.html

Qian, Yong, Thomas J. Butler, Krista Opsahl-Ong, Nicholas S. Giroux, Carlo Sidore, Ramaiah Nagaraja, Francesco Cucca, et al. 2017. “fastMitoCalc: An Ultra-Fast Program to Estimate Mitochondrial DNA Copy Number from Whole-Genome Sequences.” Bioinformatics 33 (9): 1399–1401.

Useage

Whole Genome Sequencing

bsub -J "mDNACN" -P acc_load -q premium -n 2 -R span[hosts=1] -R rusage[mem=16000] -W 01:00 -L /bin/bash -o data/mDNACN.stdout -eo data/mDNACN.stderr \
"perl src/fastMitoCalc/fastMitoCalc.pl -f raw/MAP15387421.final.bam -w data -p src/fastMitoCalc/BaseCoverage"

src/fastMitoCalc/fastMitoCalc.pl -f raw/MAP15387421.final.bam -w data -p src/fastMitoCalc/BaseCoverage

Whole Exome Sequencing

If sequence reads from the mtDNA are not targeted, need to use off target reads only. The requires .bed file of the targeted captures regions for the scpefic exome kit. For example the IDT xGen Exome Research Panel from the UKB.

Extract all off target reads

bedtools intersect -v -abam the.bam -b the.bed -bed | awk '{print $1, $2, $3}' OFS="\t" > test.bed

Extract all off target reads that 50kb away from target reads

• extract off target reads
• write out chr, start, end
• remove values of -1
• remove off target reads within 50kb

bedtools intersect -v -abam WES.bam -b Targets.bed -bed | \
awk '{print $1, $2, $3}' OFS="\t" | awk '$2!=-1' | \
bedtools window -w 50000 -v -a stdin -b Targets.bed > OffTargets.bed

Subset the number of reads from the .bed file to reduce computation time

• Select reads maping to autosomes only
• calculat length of reads and remove any with less then 100bp
• randomly sample 2% of reads from each chromosome
• Write out

read_table2('OffTargets.bed', col_names = F, col_types = list(X1 = col_character())) %>%
  filter(X1 %in% c(1:22)) %>%
  mutate(length = X3 - X2) %>%
  filter(length >= 100) %>%
  group_by(X1) %>%
  sample_frac(0.02, replace = F) %>%
  ungroup() %>%
  write_tsv('OffTargetsClean.bed', col_names = F)

fastMitoCalc limited to the regions in the bed file only

perl scr/fastMitoCalc.pl -f WES.bam -b OffTargetsClean.bed -w data -p src/BaseCoverage