amplicon)sequencing)using)next gen)technology) - … · amplicon)sequencing)using)next...

Post on 23-Apr-2018

234 Views

Category:

Documents

3 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Amplicon  sequencing  using  next-­‐gen  technology  

STAMPS  course  2013  Hilary  Morrison  

&  Sharon  Grim,  Joe  Vineis  

Overview  •  Tag  sequencing    •  Primer  design  •  Amplicon  sequencing  on  two  Illumina  plaIorms  –  fusion  primers  – opKmizaKon  – piIalls  – current  protocols  

WHAT  YOU  WILL  NOT  HEAR  ABOUT  

RETIRED

EXCHANGED

Targets  

•  Ribosomal  RNA  (SSU,  LSU)  •  Internal  transcribed  spacers  (ITS)—fungi  •  FuncKonal  genes  – nirS  – butyrate  kinase,  butyrate  transferase  – others  

Which  variable  regions  to  target?  

V1   V3   V7   V8   V9  

8F  

V2   V4   V5   V6  

V6                (967F-­‐1046R)                  60  bp  V4                (518F-­‐806R)                      288  bp  V4V5      (518F-­‐1064R)                  550  bp  V9                (1380/1389-­‐1513)  for  eukaryotes    

341   518   806   926   967  1064   1513  1380  

ITS  (fungi)  

18S  

ITS1            ~350-­‐400  bp  ITS2            ~250-­‐300  ITS1-­‐ITS2  ~550-­‐800  

ITS1F   ITS1R/ITS2F   ITS2R  

ITS1   ITS2   28S  5.8S  

AmplificaKon  process  •  DOs:  

–  High  quality  DNA  template:  e.g.  260/280  ~2  –  Replicate  reacKons  using  high-­‐fidelity  polymerase  –  NegaKve  controls  –  Pool,  clean,  size-­‐select  

•  Qiagen  MinElute  (FLX;  Illumina  v6),    •  Ampure  (Titanium;  Illumina  v4v5);    •  Pippin  Prep  (Illumina  pools)  

–  QuanKtaKon  •  e.g.  PicoGreen,  qPCR    

–  VisualizaKon  •  Bioanalyzer,  Caliper,  gel  

•  DON’Ts:  –   MDA  or  nested  amplificaKon  

Pre-­‐Ampure   Post-­‐Ampure  

What  can  go  wrong?  (sources  of  error)  •  ContaminaKon  •  PCR  error  (enzyme  misincorporaKon)  •  Chimeras  •  PCR  ectopic  amplificaKon  (non-­‐rRNA)  •  MulKple  Sequence  Alignment  •  Distance  measurements  (gaps,  etc.)  •  Clustering  methods  

Post  sequencing  •  Basecalling  •  QC  pipeline    •  Taxonomy  assignment  or  clustering  – QIIME    – GAST  (Sue  Huse,  8/4)  – mothur    –  RDP  Pipeline  (hjp://rdp.cme.msu.edu/)  – Oligotyping  –  basic  BLAST  

•  VAMPS  

16S-­‐SPECIFIC  PRIMER  DESIGN  

Design  primer  suite  for  conserved  sites  

•  Start  with  known  primers  •  Check  matches  against  SILVA  or  other  rRNA  database  – Phyla  missed?  – ProporKon  missed?  

•  Add  variants  or  degenerate  sites  to  expand  coverage  

Literature  review  

Database  sources  

3,194,778    Total        739,633      SSURef  

Example:  bacterial-­‐archaeal  universal  primer  v.  1  

0.0%  

10.0%  

20.0%  

30.0%  

40.0%  

50.0%  

60.0%  

70.0%  

80.0%  

90.0%  

100.0%  

0  mismatches   1  mismatch   2  mismatches   3  mismatches  

Bacteria  

Archaea  

Eukarya  

GGACTAC[ACG][CG]GGGTATCTAAT    

Example:  bacterial-­‐archaeal  universal  primer  v.2  

0.0%  

10.0%  

20.0%  

30.0%  

40.0%  

50.0%  

60.0%  

70.0%  

80.0%  

90.0%  

100.0%  

0  mismatches   1  mismatch   2  mismatches   3  mismatches  

Bacteria  

Archaea  

Eukarya  

GGACTAC[ACG][CG]GGGTATCTAAT     GGACTAC[ACT][ACG]GGGT[AT]TCTAAT    

0.0%  

10.0%  

20.0%  

30.0%  

40.0%  

50.0%  

60.0%  

70.0%  

80.0%  

90.0%  

100.0%  

0  mismatches   1  mismatch   2  mismatches   3  mismatches  

Bacteria  

Archaea  

Eukarya  

V6  primers  967F  is  pool  of  4  oligos  with  2-­‐16X  degeneracy  

CNACGCGAAGAACCTTANC CAACGCGMARAACCTTACC ATACGCGARGAACCTTACC CTAACCGANGAACCTYACC

1046R:  4  oligos  or  1  oligo  with  16X  degeneracy  CGACAGCCATGCANCACCT CGACAACCATGCANCACCT CGACGGCCATGCANCACCT CGACGACCATGCANCACCT

CGACRRCCATGCANCACCT

V6  Primer   Sequence   Target   RefDB  exact  match  

967F1-­‐PP   CNACGCGAAGAACCTTANC   Bacteria   63.3%  

967F-­‐AQ   CTAACCGANGAACCTYACC   Deferribacteres,  Aquificales   0.1%  

967F-­‐U12   CAACGCGMARAACCTTACC   AddiKonal  Proteobacteria   14.0%  

967F-­‐U3   ATACGCGARGAACCTTACC   AddiKonal  Bacteroides;  Spirochaetes   12.7%  

1046R   CGACRRCCATGCANCACCT   Bacteria   95.4%  

Keep  level  of  degeneracy  low  (2-­‐16X)  Are  rare,  but  important  groups  missed?  Primers  with  1-­‐2  mismatches  open  work—pay  ajenKon  to  Tm  

V4V5  Primer   Sequence   Target   RefDB  exact  match  

518F   CCAGCAGCYGCGGTAAN   Bacteria   97.0%  

926R1   CCGTCAATTCNTTTRAGT   Bacteria   96.2%  

926R3   CCGTCAATTTCTTTGAGT  Verrucomicrobia,  

Delta  proteobacteria,  AcKnobacteria  

2.3%  

926R4   CCGTCTATTCCTTTGANT   Epsilon  proteobacteria   1.3%  

Why  not  use  CCGTCWATTYNTTTRANT?  

128  variants  and  only  40  match  a  known  bacterial  sequence!      

Fusion  primer  modificaKons  •  Phosphorothioate  linkages  •  Barcoding:  5-­‐12  nt  •  Indexing:  6-­‐12  nt  •  Fusion  with  plaIorm-­‐specific  sequencing  primer  sequences  (vs.  ligaKon)  

•  Illumina  fusion  primers  include  bridge  adapter  regions  

ILLUMINA  TAG  SEQUENCING  

Illumina  amplicon  strategy  

•  Fused  primer  approach,  not  adapter  ligaKon  – Primers  much  longer  than  454  versions  –  Include  bridge  adapter  sequences  – Minimize  cost  of  barcoded  primer  sets  

•  Many  labs  now  have  Knight  515F/806R  sets                                  (V4;  >1000  indices)  •  CombinaKon  of  12  indices  x  8  inline  barcodes  creates  96-­‐plex  set  for  other  targets  

Illumina  amplicon  targets  

•  How  long  are  the  reads?  – Hiseq:  100/150  PE  – Miseq  250  PE  and  longer  

•  How  good  is  the  quality  towards  the  end?  – Undetected  errors  will  inflate  alpha-­‐diversity  – Merged  reads  require  overlap  

100  nt  PE  HiSeq1000  

V6R:  1046R  

~60  bp  complete  overlap  

967F    

Read  1    

Read  2      

Longer  reads  on  MiSeq  

V5R:  926R  

50-­‐100  nt  overlap  

V4F:  518F    

Read  1    

Read  2      

Start  with  v6  

•  Original  454  pyrotag  target-­‐lots  of  experience  •  Sequence  overlap  with  v6v4  datasets  •  Cheaper  to  do  •  Greater  sampling  depth  – 300-­‐400  million  clusters  per  lane  possible  – 3M  reads  at  96-­‐plex  – 300K  reads  at  960-­‐plex  

rRNA…TAATCTWTGGGVHCATCAGG-CCGACTGACTGA-TTGCGTGCGATC-TAGAGCATACGGCAGAAGACGAAC!

5' AATGATACGGCGACCACCGAGATCTACAC-TATGGTAATTGT-GTGCCAGCMGCCGCGGTAA…rRNA!

5’  adapter   “pad/linker”*  

*designed  to  have  no  homology  to  16s  or  Illumina  adapters/primers  

“515F”  

Knight  lab  sequencing  primer  1  

“806R-­‐rc”   “pad/linker-­‐rc”*   index-­‐rc   3’  adapter-­‐rc  

Knight  lab  sequencing  primer  2  RC  is  indexing  sequencing  primer  

Fusion  primer  design  

Learned  during  Miseq  Training    

•  First  4  bases  of  read  1  used  for  cluster  finding  (no  change)  

•  Next  8  bases  (up  through  base  12)  need  to  be  high-­‐complexity  because  of  phasing/prephasing  calculaKon  

•  This  is  also  true  of  read  2  •  Thus,  need  to  have  randomness  in  cycles  1-­‐12  of  both  reads!  

In-­‐line  bar  codes  vs.  indices    •  In-­‐line  bar  codes  cut  into  sequencing  reads,  but:  •  Cluster  finding  requires  random  distribuKon  of  A,C,G,T  at  

start  of  read  1  •  We  designed  our  primers  to  provide  this  diversity  

–  5’  NNNN  (1-­‐4)  –  Careful  mix  of  bar  codes  (5-­‐9)  

•  Didn’t  work  –  Clusters  are  found,  but  amplicons  are  sKll  ‘low  complexity’  samples  

–  Diversity  needed  at  first  12  bases  of  both  read  1  and  read  2  •  Had  to  add  high  proporKon  of  PhiX  or  other  ‘random’  

library  

5’  AATGATACGGCGACCACCGAGATCTACAC-­‐  TCTTTCCCTACACGACGCTCTTCCGATCT-­‐  NNNNTCAGC-­‐  CNACGCGAAGAACCTTANC  3’  

967F_PP_N4TCAGC:  one  of  4  primers  with  TCAGC  barcode;  forward  read  

Fusion  primer  design  

5’  CAAGCAGAAGACGGCATACGAGAT-­‐  CGTGAT-­‐  GTGACTGGAGTTCAGACGTGTGCTCTTCCGATCT-­‐  CGACRRCCATGCANCACCT  3’  

1046R_index_1  :  single  degenerate  primer  with  index;  reverse  read  

Bridge  adapter  Sequencing  primer  binding  In-­‐line  barcode  16S  PCR  

Bridge  adapter  Index  Sequencing  primer  binding  (x2)  16S  PCR  

Similar  for  v4v5  suite  

Primer  synthesis  •  >80  nt  long  •  PurificaKon  needed  –  DesalKng  only  –  PAGE  (used  for  first  set;  expensive)  –  No  other  opKon  because  of  oligo  length  

•  Pooled  4  967F  primers  for  each  barcode  (V6)  •  Pooled  3  926R  primers  for  each  index  (V4V5)  •  Final  set  of  8  forward/barcoded  and  12  reverse/indexed  primers  =  96  unique  combinaKons  for  each  domain  

•  Working  stocks  @10  uM  each  

ILLUMINA  SEQUENCING  

Effect  of  low-­‐complexity  libraries  

•  Clusters  found  correctly  •  Then  read  hits  highly  

conserved  sequence    •  Quality  scores  reduced  •  MiKgate  with  lower  cluster  

density  and  HC  library  spike-­‐in  –  over  50%  PhiX  on  early  Miseq  runs  

•  Illumina  has  sopware  fix  for  problem  on  Miseq  –  now  ~2%  PhiX  on  Miseq  runs  

Control  gDNA:  known  community  

Error  rate  

•  Generated  data  from  HMP  mock  community  gDNA  •  Expected  sequences  recovered  •  Aligned  unique  sequences  to  reference  •  Sequencing  errors,  not  quality  scores  

• Overall  error:  0.52%  

How  does  Illumina  tag  sequencing  compare  to  pyrotag  sequencing?  

Analysis  of  v6  data  from  real  samples  

Sandra  Mclellan  samples:  Illumina  (HiSeq)  overrepresents  AcKnomyces;  misses  Arcobacter  almost  enKrely  

Samples  are:  3  HiSeq  v6  amplicon  datasets  and  2  454  datasets  (v6v4  and  v6  only)  

Does  length  of  primers  bias  recovery?  

Test  alternaKves  in  2-­‐step  PCR.  – 16s  primers  only,  followed  by  full-­‐length  fusion  – 16s  plus  part  of  Illumina  adapters,  followed  by  adapters  to  bridge  

Sample,  condidons   Barcode   Index  

SS_WWTP_1_25_11,  16s  only  20  cycles   GAGAC   7:  GATCTG/CAGATC  

SS_WWTP_4_11_11,  16s  only  20  cycles   ACGCA   8:  TCAAGT/ACTTGA  

SS_WWTP_1_25_11,  split  primers     TCAGC   1:CGTGAT  

SS_WWTP_4_11_11,  split  primers     TCAGC   2:ACATCG  

“Split”  primers  

967F_PP+seq CACGACGCTCTTCCGATCTACGTTCAGCCNACGCGAAGAACCTTANC

967F_AQ+seq CACGACGCTCTTCCGATCTCGTATCAGCCTAACCGANGAACCTYACC

967F_U2+seq CACGACGCTCTTCCGATCTGTACTCAGCCAACGCGMARAACCTTACC

967F_U3+seq CACGACGCTCTTCCGATCTTACGTCAGCATACGCGARGAACCTTACC

ACGTbridge+seq AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTACGT

CGTAbridge+seq AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTCGTA

GTACbridge+seq AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTGTAC

TACGbridge+seq AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTTACG

1046R+seq GGAGTTCAGACGTGTGCTCTTCCGATCTCGACRRCCATGCANCACCT

BridgeIndex_1 CAAGCAGAAGACGGCATACGAGATCGTGATGTGACTGGAGTTCAGACGTGTGCTCTTCC

BridgeIndex_2 CAAGCAGAAGACGGCATACGAGATACATCGGTGACTGGAGTTCAGACGTGTGCTCTTCC

PCR  condiKons  Standard  3  x  33  ul  amplicon  reacKon  plus  negaKve  control  1st  round:  either  16s  only  primers  or  16s+parKal  adapter  Only  20  cycles  (vs.  30)  Pooled  3  reacKons  Diluted  10  ul  to  500  (1:50)  Used  1  ul  diluKon  in  2nd  round  of  PCR  with  2nd  set  of  primers,  5  cycles  

PCR  results  

16s  only,  20  cycles  16s  only,  +5  cycles  16s  only,    negaKve  16s  only,  20  cycles  16s  only,  +5  cycles  16s  only,    negaKve  Split,  20  cycles  Split,  +5  cycles  Split,    negaKve  Split,  20  cycles  Split,  +5  cycles  Split,    negaKve        

Sequencing  results  

Searched  for  match  in  raw  data,  read  1,  each  dataset    >gi|343201492  Arcobacter  cibarius  TGGACTTGACATAGTAAGAACTTTCTAGAGATAGATTGGTGTCTGCTTGCAGAAACTTATATAC  

Sample,  condidons   Count  read1   Count  Arcobacter   %  

SS_WWTP_1_25_11_Bv6v4   35,612   5,525   15.50%  

SS_WWTP_1_25_11_ACGCA   214,950   2,309   1.10%  

SS_WWTP_1_25_11_GACTC   240,520   2,606   1.10%  

SS_WWTP_1_25_11_GTATC   392,570   5,445   1.40%  

SS_WWTP_1_25_11_16s   6,106,654   900,824   14.80%  

SS_WWTP_1_25_11_split   12,782,547   1,558,392   12.20%  

SS_WWTP_4_11_11_16s   5,779,659   824,789   14.30%  

SS_WWTP_4_11_11_split   17,102,934   2,108,534   12.30%  

0"

5"

10"

15"

20"

25"

30"

35"

40"

45"

Beijerinckiaceae"

Alicyclobacillaceae"

Helicobacteraceae"

Xiphinematobacteriaceae"

Crenotrichaceae"

Rhodobacteraceae"

Auran@monadaceae"

Acetobacteraceae"

Xanthobacteraceae"

Tracheophyta"

Erythrobacteraceae"

Enterococcaceae"

Lactobacillaceae"

Phyllobacteriaceae"

Neisseriaceae"

Coxiellaceae"

Sinobacteraceae"

Listeriaceae"

Clostridiaceae"

Chi@nophagaceae"

RickeJsiaceae"

Bacteroidaceae"

Williamsiaceae"

Staphylococcaceae"

Rhodocyclaceae"

Streptococcaceae"

Nocardiaceae"

Nocardioidaceae"

Burkholderiaceae"

Bdellovibrionaceae"

Methylophilaceae"

Family"I"

Bacillaceae"

Brucellaceae"

Cryomorphaceae"

Microbacteriaceae"

Planctomycetaceae"

Bradyrhizobiaceae"

Moraxellaceae"

Legionellaceae"

Hyphomicrobiaceae"

Methylobacteriaceae"

Unassigned"

Rhodospirillaceae"

Oxalobacteraceae"

Paenibacillaceae"

Alcaligenaceae"

Caulobacteraceae"

Verrucomicrobiaceae"

Comamonadaceae"

Xanthomonadaceae"

Sphingobacteriaceae"

Sphingomonadaceae"

Rhizobiaceae"

Enterobacteriaceae"

Pseudomonadaceae"

Cytophagaceae"

Flavobacteriaceae"

MTG$vs$v4v5$

M51.MTG.RC"

M51.MTG.UC"

M51.v4v5.1step"

M51.v4v5.2step"

0"

5"

10"

15"

20"

25"

30"

35"

Micromonosporaceae"

Xanthobacteraceae"

Streptomycetaceae"

Nitrosomonadaceae"

Hyphomonadaceae"

Pseudonocardiaceae"

Methylocystaceae"

Hydrogenophilaceae"

Micrococcaceae"

Rhodospirillaceae"

Rhodocyclaceae"

SAR116"

Beijerinckiaceae"

Unassigned"

PiscirickeFsiaceae"

Paenibacillaceae"

Patulibacteraceae"

Oxalobacteraceae"

Psychromonadaceae"

Bacillaceae"

Erythrobacteraceae"

Acetobacteraceae"

Nocardioidaceae"

Bartonellaceae"

Phyllobacteriaceae"

Hyphomicrobiaceae"

Francisellaceae"

Bradyrhizobiaceae"

Neisseriaceae"

Mycobacteriaceae"

Alteromonadaceae"

Methylobacteriaceae"

Brucellaceae"

Enterobacteriaceae"

Cytophagaceae"

Verrucomicrobiaceae"

Microbacteriaceae"

Caulobacteraceae"

Mycoplasmataceae"

Alcaligenaceae"

Rhizobiaceae"

Xanthomonadaceae"

Nocardiaceae"

Sphingomonadaceae"

Sphingobacteriaceae"

Flavobacteriaceae"

Comamonadaceae"

Moraxellaceae"

Pseudomonadaceae"

MTG$vs$v4v6$

50ng.8cyc.R2"%"

367.v4v6"%"

Other  evidence  for  ‘missing’  taxa  

Arcobacter  

Epsilons  Rikenella  

Primer   Tm  at  1.5-­‐3.6mM  Mg2+  

Concentradon  in  working  mix  (uM)  

Specific  targets?  

518F   64 10 Universal 926R1   52 8 Rikenella 926R3   57 1 926R4   55 1 Epsilons

Bacterial  v4v5  

Thermocycling:  94˚C,  3:00  25  cycles  of:  [94˚C,  0:30;  55-­‐60˚C,  0:45;  72˚C,  1:00]  72˚C,  2:00  4˚C,  hold  

454  v6v4  and  MiSeq  v4v5  datasets  

Sewage 19_SS 1 step 2 step Bv6v4

20_SS Bv6v4 21_SS Bv6v4

Leech I_GS1_R2 Bv6v4 REA_HDF Bv6v4

OPTIMIZING  PCR  CONDITIONS  

or,  how  Sharon  spent  her  spring  

CondiKons  to  try  

•  Primer  concentraKon  – 0.2-­‐0.4  uM  

•  Magnesium  concentraKon  – 1.5mM  –  3.67mM  

•  Annealing  temperature  – Tm  of  primers  suggest  anywhere  from  52˚C-­‐60˚C  

•  Keep  primer  composiKon  and  cycle  number  constant  

Leech  

Ladd

er  

sewage  

Ladd

er  

REA_HDF  

55˚C   57˚C   60˚C   55˚C   57˚C   60˚C   55˚C   57˚C   60˚C  

0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3  

A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B  

Effect  on  amplificaNon  of  

Condidon   Leech   REA_HDF  

Increase  temperature   Decrease     Decrease  

Increase  primer   Increase   Increase  

Increase  magnesium   Increase   Increase  

Annealing  temperature,  Primers,  A:  1.5  mM  MgSO4  B:  3.6  mM  MgSO4  All  NTC  at  55˚C  

19_SS  

Ladd

er  

20_SS  

Ladd

er  

21_SS  

55˚C   57˚C   60˚C   55˚C   57˚C   60˚C   55˚C   57˚C   60˚C  

0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3  

A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B  

Effect  on  amplificaNon  of  

Condidon   19_SS   20_SS   21_SS  

Increase  temperature   Decrease   Decrease   Decrease  

Increase  primer   Increase   Increase   Increase  

Increase  magnesium   Increase   Increase   Increase  

Annealing  temperature,  Primers,  A:  1.5  mM  MgSO4  B:  3.6  mM  MgSO4  All  NTC  at  55˚C  

(+)/(-­‐)  

55˚C   57˚C  

0.2   0.3   0.2   0.3  

A  B  A  B  A  B  A  B  

Arco_1  

Ladd

er  

Arco_3  

Ladd

er  

209_5M  

55˚C   57˚C   60˚C   55˚C   57˚C   60˚C   55˚C   57˚C   60˚C  

0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3  

A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B   A  B  A  B  A  B  A  B  A  B  A  B  

Effect  on  amplificaNon  of  

Condidon   Arco_1   Arco_3   209_5M  

Increase  temperature   Decrease   Decrease   N/A  

Increase  primer   No  Effect   Minor  increase   N/A  

Increase  magnesium   Increase   Increase   N/A  

Annealing  temperature,  Primers,  A:  1.5  mM  MgSO4  B:  3.6  mM  MgSO4  All  NTC  at  55˚C  

(+)/(-­‐)  

55˚C   57˚C  

0.2   0.3   0.2   0.3  

A  B  A  B  A  B  A  B  

Archaeal  v4v5  •  Try  annealing  temperature,  primer  concentraKon,  magnesium  

concentraKon  •  Hold  constant  1X  HiFi  Buffer,  0.1  U  Taq/uL,  template  •  Use  Av6  and  Bv6  as  posiKve  control  for  template  

Primer   Tm  at  1.5-­‐3.6mM  Mg2+  

Concentradon  in  working  mix  (uM)  

arc517F1   63 10 arc517F2   59 10 arc517F3   58 10 arc517F4   56 10 arc517F5   56 10 arc958R   57 10

Thermocycling:  94˚C,  3:00  30  cycles  of:  [94˚C,  0:30;  55-­‐60˚C,  0:45;  72˚C,  1:00]  72˚C,  2:00  4˚C,  hold  

Av4v5  

Ladd

er2  

Av6   Bv6*  

55˚C   60˚C  

NTC

 

55˚C   60˚C  

NTC

 

60˚C  

NTC

 

0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3   0.2   0.3  

A   B   A   B   A   B   A   B   A   B   A   B   A   B   A   B   A   B   A   B  

Annealing  temperature,  Primers,  MgSO4  A:  1.5  mM  MgSO4  B:  3.6  mM  MgSO4  1X  HiFi  Buffer,  0.1  U  Taq/uL.  All  NTC  at  0.2  uM  each  primer,  1.5  mM  MgCl2,  55˚C  

Effect  on  amplificaNon  of  

Condidon   Av4v5   Av6   Bv6*  

Increase  temperature   Decrease     No  change   N/A  

Increase  primer   Increase   Increase   Increase  

Increase  magnesium   Increase   Increase   Increase  

Thermocycling:  94˚C,  3:00  30  cycles  of:  [94˚C,  0:30;  55-­‐60˚C,  0:45;  72˚C,  1:00]  72˚C,  2:00  4˚C,  hold  

*In  comparison,  the  usual  cocktail  has    1X  HiFi  Buffer  2  mM  MgSO4  0.2  mM  each  dNTP  0.4  uM  each  primer  0.02  U/uL  Taq  Plenty  of  DNA  In  33  uL  total  

Arcobacter  

Epsilons  Rikenella  

Primer   Tm  at  1.5-­‐3.6mM  Mg2+   Specific  targets?  

518F   64 Universal 926R1   52 Rikenella 926R3   57 926R4   55 Epsilons

Component 454 Bv4v5 Bv6 Av6 Av4v5

X  HiFi 1 1 1 1 1

mM  MgSO4 3.67 2 2 1.5 1.5

mM  dNTPs 0.2 0.2 0.2 0.2 0.2

U/uL  Taq 0.1 0.1 0.02 0.1 0.1

uM  each  primer  (set) 0.3 0.4 0.2 0.3 0.3 annealing  temp 60 60 60 60 55

cycles 30 25 25 25 30

If  you  see  your  products  amplify  with  less  Taq...  tell  us  your  secrets.  

2  mM  Mg2+,  0.02  U/ul  Taq,  0.2  uM  primers  

3.67  mM  Mg2+,  0.1  U/ul  Taq,  0.37  uM  primers  

Current  protocols  bac  v6   arc  v6   bac  v4v5   arc  v4v5  

HiFi  buffer   1X   1X   1X   1X  MgSO4   2  mM   2  mM   3  mM   3  mM  dNTPs   0.2  mM   0.2  mM   0.2  mM   0.2  mM  

combined  primers   0.2  uM   0.3  uM   0.3  uM   0.3  uM  PlaKnum  HiFi   4  units   10  units   10  units   10  units  Template   5-­‐20  ng   5-­‐20  ng   5-­‐20  ng   5-­‐20  ng  Volume   100  ul   100  ul   100  ul   100  ul  

PCR  3min   94o   94o   94o   94o  30  sec   94o   94o   94o   94o  45  sec   60o   60o   57o   57o  1  min   72o   72o   72o   72o  2  min   72o   72o   72o   72o  

Cycles,  step  1   25   30   30   30  

Pooling,  clean  up  Qiagen  MinElute  

plate  Qiagen  MinElute  

plate   Ampure,  1:1   Ampure,  1:1  

Step  2   4  ul  step  1  eluate   0   0   0  Cycles,  step  2   5   0   0   0  QuanKtaKon   Picogreen   Picogreen   Picogreen   Picogreen  Size  selecKon   Pippin  Prep   Pippin  Prep   Ampure   Ampure  

Typical  amplificaKon  results  

4900  

1900  2900  

100  

1100  

300  

500  700  

Archaeal  v4v5  pooled,  on  Caliper  GX  Bacterial  v6  pooled,  on  Caliper  GX  

Size  selecKon  

v6  amplicons  (Pippin)  

v4v5  amplicons  (Ampure)  

Illumina  amplicon  QC  

•  Filter  out  reads  based  on:  –  ChasKty  filter  –  Ns  

•  Then  ajempt  to  merge  forward  and  reverse  reads  •  Keep  only  those  that  match  perfectly  over  the  region  of  overlap  (v6)  or  have  fewer  than  3  mismatches  (v4v5)  

•  Create  consensus  using  basecall  with  higher  Qscore  •  Trim  away  adapters,  primers  •  Product  goes  into  usual  GAST  pipeline,  but  as  unique  sequences  plus  frequency  info,  not  all  reads  

2012-­‐13  amplicon  sequencing  goals  

•  Complete  454  projects  in  progress  √  •  ReKre  GS-­‐FLX  √  •  ConKnue  to  use  HS1000  for  metagenomics,  short  reads,  V6  √  •  Use  96+  mulKplexed  V6  amplicons  on  HiSeq  (paired  end  

perfect  overlaps)  √  •  Use  mulKplexed  V4  amplicons  (no  overlap)  X  •  Run  V4-­‐V5  amplicons  on  MiSeq  (parKal  overlap  for  QC  and  

merging)  √  •  ReKre  PGM  √  

Final  comments  •  Amplicon  sequencing  is  moving  target  

–  Not  as  well  supported  by  next-­‐gen  companies  as  shotgun  sequencing  –  Databases  growing  –  Sample  sources  expanding  (animal  microbiomes,  extreme  environments,  low  

biomass  targets,  marine/freshwater/sediments/rock  surfaces)  –  Read  lengths  increasing;  paired  end  reads  possible  –  Processing  s/w  –many  approaches  to  filtering,  trimming  

•  So—  –  Look  at  your  raw  data  –  Assess  expected  coverage  by  primer  sets  –  OpKmize  amplicon  protocols  and  processing  –  Evaluate  accuracy,  not  just  quality  (run  controls)  –  Don’t  hesitate  to  discard  LQ  reads  –  Expect  technological  advances  

top related