20.109(S14):Phylogenetic and primer analyses (Day7)

From OpenWetWare

(Difference between revisions)
Jump to: navigation, search
(Part 1: Bird microbiome analysis)
Line 20: Line 20:
===Part 1: Bird microbiome analysis===
===Part 1: Bird microbiome analysis===
<font color=red>
Revise Day 7 intro and/or protocol to emphasize a few key points about the analysis, particulary the alignment phase in MEGA. Prompt students to consider why “identical” species might show up on different leaves. Also add mean/max bp differences a la W/F Green report? Helps overcome differing tree scales.
Also change naming convention so there are no _ or - for potential use in Unifrac.
</font coor>

Revision as of 18:18, 10 February 2014

20.109(S14): Laboratory Fundamentals of Biological Engineering

Home        Schedule Spring 2014        Assignments       
Module 1        Module 2        Module 3              



Molecular phylogeny, or phylogenetics, is used to study relationships among organisms. The most common approach these days involves examining nucleic acid sequences or protein data from specific genetic loci; frequently the goal is to define data down to the species level. All life forms on earth trace back to a few organisms that lived billions of years ago and all share a common descent. Groups of organisms that are closely related to each other diverged from more recent shared common ancestors. Phylogeny remains one of the only effective means of describing these relationships, which can be difficult to assess by other means.

The goals of phylogenetics are to 1) reconstruct the correct genealogical relationship between organisms/genes/sequence data and 2) to estimate their divergence since sharing a common ancestor. The process of phylogenetic reconstruction relies heavily on correct comparison of the traits under question, whether it is morphological data (such as wing lengths) or sequence data. For sequence data, comparison is made by the alignment of a set of orthologous sequences, which we will do in lab from the 16s rRNA gene.

Today, we have a choice of algorithms (distance-based, neighbor-joining, parsimony, likelihood, and other) for reconstructing a phylogenetic tree that depicts the relationships among aligned sequences. A number of models for defining how the mutations between sequences (genetic substitution) are assessed are also available. Each of these methods and models has advantages and disadvantages, which are closely considered (ideally!) in any formal published phylogenetics study. In the world of microbial community analysis, a popular choice is the neighbor-joining method (Saitou and Nei, 1987), which is one of the methods that deals most accurately and consistently with large data sets. Regardless of the best method, however, the result -- a reconstructed phylogenetic tree -- has proven to be an extremely useful qualitative and often even quantitative tool for examining the relationships among organisms.

EXPAND; Also, modify protocol to have no -/_/ in names in case people want to use UniFrac.

Somewheres here? Reconnect to goal. Or earlier?: We might ask: If two microbiomes are phylogenetically different, but functionally equivalent, does that mean they will be susceptible or resistant to similar pathogens? What do differences in microbiome structure mean for a bird’s ability to carry influenza virus, microsporidia, giardia species, or other gull associated microbes?


Part 1: Bird microbiome analysis


Revise Day 7 intro and/or protocol to emphasize a few key points about the analysis, particulary the alignment phase in MEGA. Prompt students to consider why “identical” species might show up on different leaves. Also add mean/max bp differences a la W/F Green report? Helps overcome differing tree scales.

Also change naming convention so there are no _ or - for potential use in Unifrac.


You will take several steps to analyze your bird stool sequencing data, first with your partner, and ultimately across the entire class:

  • For each ### clone of yours (e.g., #716-1 through #716-8) and your partner's (e.g., #716-9 through #716-16), you will trim and combine the forward and reverse sequencing results to get one intact 16S rRNA gene sequence.
  • For each sequence, you will use BLAST to determine the closest known bacterial species to that sequence.
  • Along with your partner, you will post the sequences and a summary of the species that you found, according to a specific template.
  • You should then align all your robust sequences, up to 16 of them, in a program called MEGA, and subsequently construct a phylogenetic tree.
    • Each team in each section must post an interim or complete alignment file for one sample ###.
    • T/R section files will necessarily be incomplete.
    • W/F morning and afternoon sections will add then add their sequences to the alignment file begun for their particular ###.
  • These trees will be posted so that cross-class comparisons can be made.
    • T/R will post provisional trees to show progress.
    • W/F morning section will post interim or complete trees, depending on the sample.
    • W/F afternoon section will post complete trees that everyone can share.
  • Finally, you may compare the MA versus AK trees by inspection, as some of them will be pretty homogeneous, or you may optionally run a UniFrac analysis. (Some guidance about UniFrac will be posted later this week.) You might also want to compare composite trees for MA (up to 16x4 sequences) versus AK.

Part A: Understand possible insert orientations within vector

  1. Recall from Day 1 the sequences of the forward and reverse primers used to broadly amplify bacterial 16S rRNA gene segments:
    • Reverse: 5' ACGGGCGGTGTGTACA
  2. Based on these sequences, you might expect that your insert will always begin with "AGA" and always end with "CGT." (Draw a picture to make sure you understand why the last three bases are as they are written here.)
  3. However, in blunt-end cloning, the insert – here our PCR product – can face in either orientation. Take a moment to figure out what other basepairs you might expect to see at the beginning or end of your sequenced insert.
    • The kind of cloning we are doing is called non-directional cloning. Directional cloning is possible when, for example, two different restriction enzymes are used to create overhangs that are complementary to the vector but not to each other.

Part B: How to download a sequence

  1. The data from Genewiz is available at the company website, linked here.
  2. Choose the "Login" link and then use "astachow@mit.edu" and "be20109" to log in.
  3. At the bottom right should be a section called Recent Results. Click on More to expand it, and then click the icon under the Results column for your particular plate.
    • T/R orders were placed on x/xx, and W/F orders were placed on y/yy.
  4. The quickest way to start working with a particular sequence is to follow the "View" link under the Seq File heading. For ambiguous data, you may want to look directly at the Trace File as well.

Part C: Prepare sequences for analysis

  1. Begin by downloading this file, which contains the DNA sequence of the vector we are using in GenBank format. Open the file in ApE (A plasmid Editor, created by M. Wayne Davis at the University of Utah), which is found on your desktop. Three items of interest are highlighted: the forward priming site, the reverse priming site and the two basepairs between which your sequence should be inserted.
  2. Follow the steps below for each clone that had successful forward and reverse sequencing reactions. In cases where only one reaction was successful, briefly check whether you can locate an insert. You should also scroll down to the bottom to check if any of your failed reactions were repeated; these are noted with an "R" and in some cases worked the second time around.
  3. Paste the forward sequence of your first candidate into a new ApE file. Locate where the vector ends and the insert begins; trim away the vector.
    • While it is easiest to find the insert by doing EditFind (or Apple-F) using the base pairs right before the insert should begin, note that the string "CCC" may be mis-sequenced as "CC" or "CCCC" because long stretches of the same base (particularly Gs and Cs) are prone to error.
  4. Paste the reverse sequence of your first candidate into yet another ApE file. Immediately use EditReverse Complement to adjust the sequence, and again trim away the vector.
    • Why is it more convenient to work with the reverse complement when sequencing from the reverse direction?
  5. In ApE, use ToolsAlign Sequence to find where the forward and reverse sequences overlap. Combine them into one sequence with no repeated parts; where both forward and reverse sequence have coverage of the gene, choose whatever combination has the fewest unknown based, or Ns (ideally none!). Save this sequence as a new file called YourTeamDay-YourTeamColor_YourSampleID-"C"Candidate Number (e.g., WFA-Purple_737-C1). Use "A" for AM and "P" for PM section.
    • You may find it easiest to print out the alignment and mark up the hardcopy in order to choose where to switch from using forward to using reverse sequence. Let the base-pair numbers be your guides.
    • In pilot testing, we have run into one case in which the forward and reverse sequences have almost no overlap. It's not clear what caused this error. Before assuming that this error has struck your data, too, be sure that you reverse-complemented your reverse sequence!
  6. Finally, depending on the orientation of your insert, you may want to reverse complement the entire sequence. Use the original sequences of the forward and reverse 16S primers to guide your decision. It is important for subsequent alignment that all sequences are 5' to 3' (begin with AGA).
  7. You must now save each sequence in .txt format. If anyone can figure out how to do this task directly in ApE, let us know! Otherwise, you can copy-paste the sequence into a program such as TextEdit, choose FileSave, and in the pulldown menu select Plain Text.

Part D: Identify species from sequences

  1. The "nucleotide BLAST" alignment program can be accessed through the NCBI BLAST page or directly from this link. Follow the steps below for each clone, one at a time.
  2. Paste the sequence text that you prepared above into the "Query" box. If there were ambiguous areas of your sequencing results, these will be listed as "N" rather than "A" "T" "G" or "C" and it's fine to include Ns in the query.
  3. Under Choose Search Set, select "16S ribosomal RNA sequences (Bacteria and Archaea)" from the Database pulldown menu.
  4. Click on the BLAST button. Matches will be shown by vertical lines between the aligned sequences, while mismatches and gaps will be shown with a dash.
  5. Because this gene is highly conserved, a number of species should come up as highly matched. However, one should (usually) be a best choice. Think carefully here rather than blindly accepting the top species listed.
    • For example, if a partial sequence for species A comes up as the top choice, a full sequence for species B comes up as the second choice, and a full sequence for species A is the third most closely matched choice, is species A or B truly closer to your original sequence?
  6. When you have decided which is best, use the linked template to document this strain and its accession number, its associated max score, query coverage, max identity, gaps, mismatches, and full taxonomy; write down these parameters for the second most closely matched species as well. The taxonomy information can be found by clicking on the accession number and looking under the "organism" heading.
    • Taxonomy order is kingdom, phylum, class, order, family, genus, and species.
  7. When a particular clone is very closely matched to two different species, you might choose to define it at a higher order, such as genus or family. When a particular clone is not well-matched to any known species (perhaps representing an unidentified or undocumented species), you might also choose to define it at a higher order when submitting this information in the phylogenetics program.
  8. Be sure to rename the candidates according to your section day, team color, and clone number.
  9. Please post all of your .txt files (up to 16) and also your Excel file to the table on today's Talk page when you have finished.

Part E: Align sequences and construct tree

For this next part you will use freely available software called Molecular Evolutionary Genetics analysis, or MEGA. Feel free to read additional information about this software at the MEGA website. What you need should already be downloaded on your laboratory computers, or you can download onto your personal computers if you wish.

Important note for W/F sections: When adding your sequences to an existing T/R alignment file, you may need to first Insert Blank Sequence, and then copy-paste into that slot.

  1. Open MEGA. In the upper left corner, click on the icon labeled Align, and choose Edit/Build Alignment from the pulldown menu. This selection should open the Alignment Explorer. When you are prompted, choose "DNA" alignment of course.
  2. Under Edit, choose Insert Sequence from File and select your first .txt file. It should appear in the explorer.
  3. Double-click to rename according to the species. Note that each sequence must have a unique name. Thus, it is best that you name according to both species and clone: for example, "Klebsiella oxytoca (TR-Blu-4). This approach will also allow us to track which sequences came from which individual preps, which might be useful information.
    • Please use the following 3-letter abbreviations for your colors: Red, Org, Ylw, Grn, Blu, Pnk, Prp, (Sil?).
  4. When you have input all sequences from your team (up to 16), choose EditSelect All, followed by AlignmentAlign by Clustal-W.
  5. Now choose DataSave Session and name the alignment according to section and clone (such as "TR-716-alignment"). Post this file on today's Talk page. That way W/F data can readily be combined with T/R data into one-tree, per gull sample, by using Open Saved Alignment Session followed by copy and paste.
    • To be clear, T/R will post 16-sequence alignment files, and W/F sections will post 32-sequence (or more) alignment files. Ditto for trees.
  6. Under Data, choose Phylogenetic Analysis.
    • When prompted, should you answer that the DNA is protein-coding or not protein-coding?
  7. Now leave Alignment Explorer and go back to the original MEGA window.
  8. From the Phylogeny icon pulldown menu, select Construct/Test Neighbor-Joining Tree. To proceed, click on Compute.
  9. Finally, choose ImageSave as PDF File to document your tree. Save according to section and sample number as before.
  10. Please post the trees on today's Talk page, so we can see that everyone had the chance to do tree analysis on their own.
    • T/R section will post trees representing T/R data only.
    • W/F section will post trees that include both T/R and W/F sequences.
  11. In larger groups, for example at office hours, we can construct trees wherein all data from a given region are combined.

Part F: Compare sets of trees

Guidance to come in a few days...

Part 2: Microsporidia primer analysis

The X PCR samples were labeled numerically, and the associated sample definitions are included in the attached file. Each group should have three consecutive reactions; note that #1-12 are reference samples (with V1-PMP2 primer set) that will be added to the gel by the teaching faculty.

Each gel will have the following general structure for sample loading:

Lane Sample (20 μL) Lane Sample (20 μL)
1 Group 1, sample 1 6 V1-PMP2, sample 2
2 Group 1, sample 2 7 V1-PMP2, sample 3
3 Group 1, sample 3 8 Group 2, sample 1
4 DNA ladder (load 10 μL) 9 Group 2, sample 2
5 V1-PMP2, sample 1 10 Group 2, sample 2

REVISE It's essential that the correct reference sample be run on each gel, and therefore that groups requiring the same reference sample pair up. Let the table below be your guide:

Gel number Reference samples Group 1 Group 2
T/R 1 Specificity (VC, EH, mixture) Red Orange
T/R 2 Specificity (VC, EH, mixture) Blue Purple
T/R 3 Sensitivity (VC: lo, mid, hi) Yellow Green
T/R 4 Sensitivity (VC: lo, mid, hi) Pink Plat runs W/F Red!
W/F 1 Specificity (VC, EH, mixture) Orange Green
W/F 2 Specificity (VC, EH, mixture) Blue Pink
W/F 3 Sensitivity (EH: lo, mid, hi) Yellow Purple
W/F 4 Sensitivity (EH: lo, mid, hi) Platinum Red runs T/R Platinum!

Sample preparation: mix by pipetting, take 20 μL, add 4 μL loading dye, then load 22 μL onto gel.

For next time

Some of you have journal clubs next time. No other required homework is due on Day 8.

  1. The following bonus assignment may be submitted on Day 8: Prepare a figure and caption for your primer design summary that shows your raw PCR results -- the agarose gel. (Later you might decide to process this data in some way, but not necessarily.) Write an early draft of the accompanying main text paragraph.

Reagent list

  • Mostly your brains!
  • Agarose gels
    • 2:1 mixture of high-resolution:standard agarose
    • Prepared in TAE buffer
    • With SYBR Safe stain (Invitrogen)
      • used at manufacturer's recommended concentration, 10000-fold dilution
  • NEB loading dye (6X stock)
  • Gels made and run in 1X TAE buffer
    • 40 mM Tris
    • 20 mM Acetic Acid
    • 1 mM EDTA, pH 8.3
  • 100 bp DNA ladder from New England BioLabs

Navigation Links

Next Day: Journal club II

Previous Day: Journal club I
Personal tools