Reconstruction of the Intra-Host Evolution of HCV Mathieu Flinders - - PowerPoint PPT Presentation

▶
reconstruction of the intra host evolution of hcv
SMART_READER_LITE
LIVE PREVIEW

Reconstruction of the Intra-Host Evolution of HCV Mathieu Flinders - - PowerPoint PPT Presentation

Oct 12, 2022 284 likes •794 views

Reconstruction of the Intra-Host Evolution of HCV Mathieu Flinders Max Planck Institute for Informatics April 5, 2013 Data Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt 2/19 April 5, 2013 Data 3 patients

slide-1
SLIDE 1

Reconstruction of the Intra-Host Evolution of HCV

Mathieu Flinders

Max Planck Institute for Informatics April 5, 2013

slide-2
SLIDE 2

April 5, 2013 2/19

Data

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-3
SLIDE 3

April 5, 2013 3/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-4
SLIDE 4

April 5, 2013 4/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

  • 3 patients (DB, FP, MR) received boceprevir (BCP) monotherapy

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-5
SLIDE 5

April 5, 2013 5/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

  • 3 patients (DB, FP, MR) received boceprevir (BCP) monotherapy
  • 3 patients (IM1, IM3, IM11) received danoprevir (DNP) monotherapy

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-6
SLIDE 6

April 5, 2013 6/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

  • 3 patients (DB, FP, MR) received boceprevir (BCP) monotherapy
  • 3 patients (IM1, IM3, IM11) received danoprevir (DNP) monotherapy
  • treatment failed to eliminate HCV

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-7
SLIDE 7

April 5, 2013 7/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

  • 3 patients (DB, FP, MR) received boceprevir (BCP) monotherapy
  • 3 patients (IM1, IM3, IM11) received danoprevir (DNP) monotherapy
  • treatment failed to eliminate HCV
  • 5–7 blood samples (A, B, C...) taken over a period of 8–50 months

(4+ years)

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-8
SLIDE 8

April 5, 2013 8/19

Data

  • 3 patients (AM/VX-8, CG/VX-1, WM/VX-9) received telaprevir (TPR)

monotherapy

  • 3 patients (DB, FP, MR) received boceprevir (BCP) monotherapy
  • 3 patients (IM1, IM3, IM11) received danoprevir (DNP) monotherapy
  • treatment failed to eliminate HCV
  • 5–7 blood samples (A, B, C...) taken over a period of 8–50 months

(4+ years)

  • HCV NS3 sequenced using Roche/454 FLX

Source: Gastroenterology Clinic, JW Goethe University Hospital, Frankfurt

slide-9
SLIDE 9

April 5, 2013 9/19

Daclatasvir (BMS-790052)

Pipeline Report (2012)

Class: NS5a inhibitor Manufacturer: Bristol-Myers Squibb Dose: 60mg, once-daily Status: phase III Population: genotypes 1–4

Danoprevir/r (RG7227)

Class: protease inhibitor (ritonavir-boosted) Manufacturer: Hoffmann-La Roche/Genentech Dose: 100mg (boosted with 100mg of ritonavir), twice-daily Status: phase III, fixed-dose combination in phase I Population: genotypes 1 and 4

slide-10
SLIDE 10

April 5, 2013 10/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

*geno2pheno[hcv]

slide-11
SLIDE 11

April 5, 2013 11/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

*geno2pheno[hcv]

slide-12
SLIDE 12

April 5, 2013 12/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

sampling time (A, B, C...) *geno2pheno[hcv]

slide-13
SLIDE 13

April 5, 2013 13/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

sampling time (A, B, C...) frequency in population (%) *geno2pheno[hcv]

slide-14
SLIDE 14

April 5, 2013 14/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

sampling time (A, B, C...) make name unique frequency in population (%) *geno2pheno[hcv]

slide-15
SLIDE 15

April 5, 2013 15/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

  • intra-patient evolutionary networks reconstructed by joining each

haplotype to its nearest ancestor: A23 B8a C9 A13 B8b C8

sampling time (A, B, C...) make name unique frequency in population (%) *geno2pheno[hcv]

slide-16
SLIDE 16

April 5, 2013 16/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

  • intra-patient evolutionary networks reconstructed by joining each

haplotype to its nearest ancestor: A23 B8a C9 A13 B8b C8

sampling time (A, B, C...) make name unique frequency in population (%) connect with nearest ancestor *geno2pheno[hcv]

slide-17
SLIDE 17

April 5, 2013 17/19

Analysis

  • region of interest: nucleotide positions 106–522 (length 417bp;

chosen to include known resistance mutations*)

  • agglomerative clustering with novel tree-cutting method for finding

haplotypes: B8a

  • intra-patient evolutionary networks reconstructed by joining each

haplotype to its nearest ancestor: A23 B8a C9 A13 B8b C8

sampling time (A, B, C...) make name unique frequency in population (%) connect with nearest ancestor dashed lines for interval > 12m *geno2pheno[hcv]

slide-18
SLIDE 18

April 5, 2013 18/19

Sampling Times

slide-19
SLIDE 19

April 5, 2013 19/19

BCP patient FP

slide-20
SLIDE 20

April 5, 2013 20/19

BCP patient MR

  • cf. Gray et al (2012)
slide-21
SLIDE 21

April 5, 2013 21/19

TPR patient CG/VX-1 (1/2)

slide-22
SLIDE 22

April 5, 2013 22/19

TPR patient CG/VX-1 (2/2)

slide-23
SLIDE 23

April 5, 2013 23/19

TPR patient WM/VX-9 (1/2)

slide-24
SLIDE 24

April 5, 2013 24/19

TPR patient WM/VX-9 (2/2)

slide-25
SLIDE 25

April 5, 2013 25/19

Conclusions

Broadly coherent structure in terms of propagation of resistance mutations and variant abundances Varied branching structures provide evidence for both strong and weak selection, consistent with the assumption that evolution will be weak

  • over short time intervals, and/or
  • in the absence of protease inhibitor,

and otherwise strong. Reversion to wild-type (i.e. the extinction of resistant strains) does

  • ccur and takes as little as 13 months

Trials of HCV protease inhibitors are potentially valuable case studies

  • f evolution with strong selection
slide-26
SLIDE 26

April 5, 2013 26/19

References

Resistance mutations: SE Schelhorn, AM Sikorski, M Zeidler, S Sierra-Aragon, B Beggel, J Büch, R Kaiser, T Lengauer, geno2pheno[hcv] (2011). Population structure: RR Gray, M Salemi, P Klenerman, OG Pybus, A New Evolutionary Model for Hepatitis C Virus Chronic Infection. PloS Pathogens (2012).

slide-27
SLIDE 27

April 5, 2013 27/19

Acknowledgments

Saarbrücken: Thomas Lengauer Glenn Lawyer Prabhav Kalaghatgi Bastian Beggel Sven-Eric Schelhorn Frankfurt: Christoph Sarrazin Stefan Zeuzem Simone Susser

slide-28
SLIDE 28

April 5, 2013 28/19

Thank you!

slide-29
SLIDE 29

April 5, 2013 29/19

DNP patient IM11 (Slide A)

slide-30
SLIDE 30

April 5, 2013 30/19

DNP patient IM11 (Slide B)

slide-31
SLIDE 31

April 5, 2013 31/19

Method

Example phylogenetic tree with N = 24 sequences:

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

slide-32
SLIDE 32

April 5, 2013 32/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

slide-33
SLIDE 33

April 5, 2013 33/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.
  • 2. Where two clusters both ≥ 3 are joined... cut!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

slide-34
SLIDE 34

April 5, 2013 34/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.
  • 2. Where two clusters both ≥ 3 are joined... cut!

Note: If number of sequences N is large, use max{3,N*5%}.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

slide-35
SLIDE 35

April 5, 2013 35/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.
  • 2. Where two clusters both ≥ 3 are joined... cut!

Note: If number of sequences N is large, use max{3,N*5%}. Q: How many clusters?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

slide-36
SLIDE 36

April 5, 2013 36/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 1+1 1+1 1+1 1+1 1+1 1+1 1+1 1+1 1+1

Step 1

all < 3

slide-37
SLIDE 37

April 5, 2013 37/19

Method

  • 1. Join sequences bottom-up, noting sizes of clusters involved.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 2 + 1 2 + 1 2 + 2 2 + 2 2 + 1 2 + 1

Step 2

all < 3

slide-38
SLIDE 38

April 5, 2013 38/19

Method

  • 2. Where two clusters both ≥ 3 are joined... cut!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3 + 3 4 + 4 3 + 1 3 + 2

Step 3

slide-39
SLIDE 39

April 5, 2013 39/19

Method

  • 2. Where two clusters both ≥ 3 are joined... cut!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3 + 3 4 + 4 3 + 1 3 + 2

Step 3 both ≥ 3

slide-40
SLIDE 40

April 5, 2013 40/19

Method

  • 2. Where two clusters both ≥ 3 are joined... cut!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3 + 3 4 + 4 3 + 1 3 + 2

Step 3 both ≥ 3

slide-41
SLIDE 41

April 5, 2013 41/19

Method

  • 2. Where two clusters both ≥ 3 are joined... cut!

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 3 + 3 4 + 4 3 + 1 3 + 2

Step 3 both ≥ 3 not both ≥ 3

slide-42
SLIDE 42

April 5, 2013 42/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 6 + 8 4 + 5

Step 4

slide-43
SLIDE 43

April 5, 2013 43/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 6 + 8 4 + 5

Step 4 both ≥ 3

slide-44
SLIDE 44

April 5, 2013 44/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 6 + 8 4 + 5

Step 4 both ≥ 3 already clustered

slide-45
SLIDE 45

April 5, 2013 45/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 14 + 9

Step 5

slide-46
SLIDE 46

April 5, 2013 46/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 14 + 9

Step 5 already clustered

slide-47
SLIDE 47

April 5, 2013 47/19

Method

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 23 + 1

Step 6 not both ≥ 3

slide-48
SLIDE 48

April 5, 2013 48/19

Method

{1, 2, 3} {4, 5, 6} {7, 8, 9, 10} {11,12,13,14} {15,16,17,18} {19,20,21,22,23} {24}* * 'star' cluster of unclustered reads

Guarantee:

  • 1. Clusters of a certain size (in this case ≥ 3; in general ≥ 5%);
  • 2. Statistical significance of average over clusters.

Final clustering

slide-49
SLIDE 49

April 5, 2013 49/19

Acknowledgments

Saarbrücken: Thomas Lengauer Glenn Lawyer Prabhav Kalaghatgi Bastian Beggel Sven-Eric Schelhorn Frankfurt: Stefan Zeuzem Christoph Sarrazin Simone Susser