Upload
others
View
5
Download
0
Embed Size (px)
Citation preview
www.pnas.org/cgi/doi/10.1073/pnas. 115
freq
uenc
y pe
r 106 C
D4+ T
cel
ls
10-3
10-2
10-1
100
101
Q2 VOA
Intact
P<0.001
wk-2 wk12
92429243924492469247924192529255
Fig. S1. Comparisons of frequencies of intact proviruses and inducible proviruses. Scatter plot showing the frequencies of intact proviruses and inducible proviruses measured by NFL sequencing and Q2VOA, respectively. Each dot represents a different participant. Horizontal bars indicate median values. Statistical significance was determined using two-tailed Mann-Whitney U test.
1813512
1.11%
0.22%
0.03%
3.01%
1.37%
0.21%
0.02%
5.36%
1.35%
0.15%
0.04%
6.36%
1.45%
0.21%
0.008%
4.51%
2.12%
0.87%
0.004%
3.05%2.03%
0.55%
0.06%
4.09%3.24%
0.21%
0.10%
9.15%
1.56%
0.26%
0.13%
4.73%
0.74%
0.08%
0.05%
2.77%
1.37%
0.11%
0.007%
3.92%
2.21%
0.36%
0.03%
5.63%
9242 9243 9244 9246 9247 9241 9252 9255 9245 9249 9253
0.001
0.01
0.1
1
10
100
norm
aliz
ed p
erce
ntag
e of
pro
viru
ses
categorygagenvfull_sizeintactinducible
Fig. S2. Comparisons of different reservoir measurements. Bar graph showing the proportion of env+ proviruses, near full size proviruses, intact proviruses, and inducible proviruses within gag+ proviruses population.
0
2
4
6
8
Q2 VOA
NFL
Ave
rage
pai
rwis
e di
stan
ce
wk-2 wk12
92429243924492469247924192529255
9245
92539249
9254
P=0.79
Fig. S3. HIV-1 env sequences diversity.Scatter plot showing HIV-1 env sequence diversity from Q2VOA or NFL sequencing measured by average pairwise nucleotide distance. Each dot represents a different participant. Horizontal bars indicate median values. Statistical significance was determined using two-tailed Mann-Whitney U test.
0.004 0.004 0.004 0.004
0.001 0.004 0.5 0.004
92469242 9244 9247 9241
92459254 9255 9249 9253
0.004
0.02
Fig. S4. Phylogenetic trees of env sequences. Maximum-likelihood phylogenetic trees of env sequences obtained from Q2VOA at the pre-infusion (green) and week 12 (blue), NFL at the pre-infusion (purple) and week 12 (orange), and rebound viruses from SGA or outgrowth cultures (red).
0 50 100 1500
50
100
150
% env clonality
% o
verla
p w
ith Q
2 VO
A 924292439244924692479241925292559245
92539249
r = 0.808P = 0.0015
Fig. S5. Correlation between shared sequences from two methods and clonality. Scatter plot showing Pearson correlation between the percentage of clones and percentage of overlap between Q2VOA and NFL sequencing. Each dot represents a different participant.
NFL
Q2 VOA
106
107
108
109
num
ber o
f CD
4+ T c
ells
test
ed
wk-2 wk12
924292439244924692479241925292559245
92539249
P<0.0001
Fig. S6. Number of CD4+ T cells tested. Scatter plot showing the number of CD4+ T cells tested in Q2VOA and NFL sequencing. Each dot represents a different participant. Horizontal bars indicate median values. Statistical significance was determined using two-tailed Mann-Whitney U test.
9242
9243
9244
9245
9246
9247
9241
9249
9252
9253
Freq
uenc
y
Number of mutations
Predicted mutation accumulationHamming distanceDistance including possibility of recombination
Fig. S7. Relationship between latent and rebound sequences.Histograms show the proportion of env sequences (y-axis) and number of mutations (x-axis). The blue bars represent the Hamming distance between rebound and NFL/Q2VOA sequences. The grey bars represent the predicted distance between rebound and NFL/Q2VOA sequences based on a simulation of mutation accumulation during the ATI period for each participant. Yellow bars represent the predicted distance between the rebound and NFL/ Q2VOA sequences when the possibility of recombination is included.
ORFs intact?
no
packaging signal/MSD intact?
intact packaging signal/MSD deletions/mutations
yes no
hypermutated?
hypermutation indel/nonsense
yes no
yes
Fig. S8. Process used to identify intact and defective proviruses. Assembled sequences were aligned to the HXB2 genome to identify premature stop codons, out-of-frame insertions or deletions (indel), or packaging signal (Ψ) and the major splice donor (MSD) site deletions and mutations.