Getting Sequences from GenBank using R-packages

• Open R and select the working directory where you want to output sequence files

Misc>Change Working Directory>select a folder (e.g., R_class_winter_2015)

Alternatively:

setwd("/Users/jcsantos/Desktop/R_class_winter_2015/1_getting_sequences_from_GenBank")!
!
#I open terminal and drag the folder to it to get the path. Then, copy and paste.!

• We need to install and load the following packages:

[Link]("ape")!
[Link]("seqinr")!
!
library(ape) #this is a general R-package for phylogenetics and comparative methods!
library("seqinr") #this is an specialized package for nucleotide sequence management!

• Let’s check that our packages have been loaded correctly

!
sessionInfo()!
34
 Getting Sequences from GenBank using R-packages

• Let's use 'ape' to read the sequence from GenBank this with the function: ?[Link]!

• This function connects to the GenBank database, and reads nucleotide sequences using
accession numbers given as arguments.

• Usage (do not run)

[Link]([Link], [Link] = [Link], [Link] = FALSE)!

#[Link]: a vector of mode character giving the accession numbers.!

#[Link]: the names to give to each sequence; by default the accession numbers.!
#[Link]: a logical whether to return the sequences as an object "DNAbin”.!

• Let's read the casque-headed lizard (Basiliscus basiliscus) RAG1 sequence JF806202
!
seq_1_DNAbin <- [Link]("JF806202") #save as DNAbin object:!
attr(seq_1_DNAbin, "species") #to get the specie name of the sequence!
seq_1_DNAbin$JF806202!
str(seq_1_DNAbin) # we get the structure of the object!
!
#save as character object:!
!
seq_1_character <- [Link]("JF806202", [Link] = TRUE)!
seq_1_character #this is not a very nice format!

35
 Read sequences using accession numbers
• Create a vector of GenBank accession numbers that we want

lizards_accession_numbers <- c("JF806202", "HM161150", "FJ356743", "JF806205", !

"JQ073190", "GU457971", "FJ356741", "JF806207",!
"JF806210", "AY662592", "AY662591", "FJ356748", !
"JN112660", "AY662594", "JN112661", "HQ876437", !
"HQ876434", "AY662590", "FJ356740", "JF806214", !
"JQ073188", "FJ356749", "JQ073189", "JF806216", !
"AY662598", "JN112653", "JF806204", "FJ356747", !
"FJ356744", "HQ876440", "JN112651", "JF806215",!
"JF806209") !
#create a vector a GenBank accession numbers!

• Get those sequences and save them in a single DNAbin object:!

!
lizards_sequences <- [Link](lizards_accession_numbers) #read sequences and place
them in a DNAbin object!
!
lizards_sequences #a brief summary of what is in the object, including base composition!
!
!
str(lizards_sequences) #a list of the DNAbin elements with length of the sequences!
#notice the one of the attributes is the species names!

36
 Read sequences and create a fasta file format

• Lets explore more the DNAbin object:

attributes(lizards_sequences) #see the list of attributes and contents !

!
names(lizards_sequences) #the accession numbers!
!
attr(lizards_sequences, "species") # we get the species list. Notice this !
# attr is slightly different function!

• However, it is hard remember which accession number corresponds to which species.

So we can use the previous information to create first a vector with such information

lizards_sequences_GenBank_IDs <- paste(attr(lizards_sequences, "species"), names

(lizards_sequences), sep ="_RAG1_") !
!
## build a character vector with the species, GenBank accession numbers, and gene!
## name "_RAG1_” this is its common abbreviation: recombination activating protein 1!
## notice the use of the paste function: textA, textB, textC!
## results in: textAtextCtextB!
!
lizards_sequences_GenBank_IDs #a more informative vector of names for our sequences!

37
 Write a fasta file format
• Let’s write sequences to a text file in fasta format using [Link](). However, only
accession numbers are included.

?[Link] # This function writes in a file a list of DNA sequences in sequential,

interleaved, or FASTA format.!
!
### we are going to write in fasta format!
!
[Link](lizards_sequences, file ="lizard_fasta_1.fasta", format = "fasta", append =
FALSE, nbcol = 6, colsep = " ", colw = 10)!
!
########### Some relevant arguments for [Link]()!
!
#x: a list or a matrix of DNA sequences.!
!
#file: a file name specified to contain our sequences!
!
#format: Three choices are possible: "interleaved", "sequential", or "fasta", or any
#unambiguous abbreviation of these.!
!
#append: a logical, if TRUE the data are appended to the file without erasing the data
#possibly existing in the file, otherwise the file is overwritten (FALSE the default).!
!
#nbcol: a numeric specifying the number of columns per row (6 by default)!
!
#colsep: a character used to separate the columns (a single space by default).!
!
#colw: a numeric specifying the number of nucleotides per column (10 by default).!
###########! 38
 Write a fasta file format
• Lets explore our recently created file ‘lizard_fasta_1.fasta’. Drag and drop this file in the
text editor

• This file has our sequences, but we only have the accession numbers
39
 Rewrite a fasta file format with more information
• Read our fasta file using the seqinr package

lizard_seq_seqinr_format <- [Link](file = "lizard_fasta_1.fasta", seqtype = "DNA",

[Link] = TRUE, forceDNAtolower = FALSE)!
!
lizard_seq_seqinr_format #this shows different form to display the same sequence !
#information !
!
• Rewrite our fasta file using the name vector that we created previously
!
[Link](sequences = lizard_seq_seqinr_format, names = lizards_sequences_GenBank_IDs,
nbchar = 10, [Link] = "lizard_seq_seqinr_format.fasta")!
!
#Suggestion: Do not rearrange, delete or add sequenced to the fasta file, as the
function will assign the names in the order provided in the file and the name vector!
!
• Let’s check our new fasta file ‘lizard_seq_seqinr_format.fasta’
!

40
 Get sequences without using accession numbers

• We can use a package that use an API (application programming interface) to interact
with the NCBI website.

More info in: [Link]

[Link] ("rentrez")!
library (rentrez)!
!
• Let’s get some lizard sequences
!
lizard <- "Basiliscus basiliscus[Organism]" #We want a character vector!
!
#nucleotide database (nuccore) and retmax determines the max number!
lizard_search <- entrez_search(db="nuccore", term=lizard, retmax=40) !
lizard_search!
lizard_search$ids #gives you the NCBI ids!
!
!
#gets your sequences as a character vector!
lizard_seqs <- entrez_fetch(db="nuccore", id=lizard_search$ids, rettype="fasta")!
lizard_seqs!

41
 Get sequences without using accession numbers

• Lets get our Basiliscus basiliscus RAG 1 sequence

!
Bbasiliscus_RAG1 <- "Basiliscus basiliscus[Organism] AND RAG1[Gene]”!
!
Bbasiliscus_RAG1_search <- entrez_search(db="nuccore", term=Bbasiliscus_RAG1, retmax=10) !
#nucleotide database (nuccore) and retmax determines no more than 10 access numbers to
return!
!
Bbasiliscus_RAG1_search$ids #gives you the NCBI ids!
!
Bbasiliscus_RAG1_seqs <- entrez_fetch(db="nuccore", id=Bbasiliscus_RAG1_search$ids,
rettype="fasta")!
!
Bbasiliscus_RAG1_seqs #notice \n (new line) delimiter. Other common delimiters are \r !
#(carriage return) and \t (tab).!
!
write(Bbasiliscus_RAG1_seqs, "Bbasiliscus_RAG1.fasta", sep="\n") #gets sequence to a
file!
!
• We can read our fasta file using seqinr package
!
Bbasiliscus_RAG1_seqinr_format <- [Link](file = "Bbasiliscus_RAG1.fasta", seqtype =
"DNA", [Link] = TRUE, forceDNAtolower = FALSE)!
!
Bbasiliscus_RAG1_seqinr_format # you can also check the .fasta file in the working
folder!
!
42
 Example: Accessing Cytochrome B Sequences

• We can use the ‘rentrez’ package to get lots of sequences using taxonomic
classifications for specific markers
!
Liolaemus_CYTB <- "Liolaemus[Organism] AND CYTB[Gene]” !
!
#This is a well-studied gene from this genus of South American lizards!
!
Liolaemus_CYTB_search <- entrez_search(db="nuccore", term=Liolaemus_CYTB, retmax=100) !
!
Liolaemus_CYTB_search #There are 2539 sequences that match this query !
!
!
• Let’s adjust the search and fetch all sequences of of sequences using taxonomic
classifications for specific markers!
!
!
Liolaemus_CYTB_search_2 <- entrez_search(db="nuccore", term=Liolaemus_CYTB, retmax=2539)!
!
Liolaemus_CYTB_search_2$ids #gives you the NCBI ids!
!
Liolaemus_CYTB_seqs <- entrez_fetch(db="nuccore", id=Liolaemus_CYTB_search_2$ids ,
rettype="fasta")!
!
#we get an error “client error: (414) Request-URI Too Long”. We are asking too many
sequences!

43
 Example: Accessing Cytochrome B Sequences

• Lets adjust the search and fetch by smaller chunks so we can get the first 1500
sequences!
!
Liolaemus_CYTB_seqs_part_1 <- entrez_fetch(db="nuccore", id=Liolaemus_CYTB_search_2$ids
[1:500] , rettype="fasta")!
!
Liolaemus_CYTB_seqs_part_2 <- entrez_fetch(db="nuccore", id=Liolaemus_CYTB_search_2$ids
[501:1000] , rettype="fasta")!
!
Liolaemus_CYTB_seqs_part_3 <- entrez_fetch(db="nuccore", id=Liolaemus_CYTB_search_2$ids
[1001:1500] , rettype="fasta")!
!
!
• Lets write as single file by appending all 3 chucks of sequences
!
write(Liolaemus_CYTB_seqs_part_1, "Liolaemus_CYTB_seqs.fasta", sep="\n")!
!
write(Liolaemus_CYTB_seqs_part_2, "Liolaemus_CYTB_seqs.fasta", sep="\n", append = TRUE)
#it gets the sequences to the same file by changing the logical argument of append from
#the default FALSE to TRUE (i.e., can abbreviate TRUE with T or other unambiguous
#abbreviation)!
!
write(Liolaemus_CYTB_seqs_part_3, "Liolaemus_CYTB_seqs.fasta", sep="\n", append = TRUE) !
#you will get a 1.3 Mb file with all 1500 sequences!

44
 Example: Accessing Cytochrome B Sequences

• We can read our fasta file using the seqinr package and rename the sequences
!
Liolaemus_CYTB_seqs_seqinr_format <- [Link](file = "Liolaemus_CYTB_seqs.fasta",
seqtype = "DNA", [Link] = TRUE, forceDNAtolower = FALSE)!
!
Liolaemus_CYTB_seqs_seqinr_format!
!
Liolaemnus_CYTB_names <- attr(Liolaemus_CYTB_seqs_seqinr_format, "name")!
!
Liolaemnus_CYTB_names <- gsub("\\..*","", Liolaemnus_CYTB_names) !
!
#eliminate characters after "." using ?gsub (Pattern Matching and Replacement)!
!
Liolaemnus_CYTB_names <- gsub("^.*\\|", "", Liolaemnus_CYTB_names) !
!
#eliminate characters before "|" using ?gsub (Pattern Matching and Replacement)!
!
Liolaemnus_CYTB_names!
!
!

45
 Example: Accessing Cytochrome B Sequences

• We can read our fasta file using ape package to get accession numbers and species
names
!
Liolaemus_CYTB_seqs_ape_format <- [Link](Liolaemnus_CYTB_names)!
!
attr(Liolaemus_CYTB_seqs_ape_format, "species") !
#to get the species names of the sequence!
!
names(Liolaemus_CYTB_seqs_ape_format)!
!
Liolaemus_CYTB_seqs_GenBank_IDs <- paste(attr(Liolaemus_CYTB_seqs_ape_format,
"species"), names(Liolaemus_CYTB_seqs_ape_format), sep="_CYTB_") !
## build a vector object with the species, GenBank accession numbers, and type of gene!
!
Liolaemus_CYTB_seqs_GenBank_IDs #vector of names to add to sequences!
!
# Read our fasta file 'Liolaemus_CYTB_seqs.fasta' using seqinr package!
!
Liolaemus_CYTB_seqs_seqinr_format <- [Link](file = "Liolaemus_CYTB_seqs.fasta",
seqtype = "DNA", [Link] = TRUE, forceDNAtolower = FALSE)!
!
# Rewrite our fasta file using the name vector that we created previously!
!
[Link](sequences = Liolaemus_CYTB_seqs_seqinr_format, names =
Liolaemus_CYTB_seqs_GenBank_IDs, nbchar = 10, [Link] =
"Liolaemus_CYTB_seqs_seqinr_format.fasta”)!

46
47
 Alignment and Simultaneous Tree Estimation

• We are going to use SATe-2 (SATé - Simultaneous Alignment and Tree Estimation)
!
!
URL: [Link]
!
!

48
 Alignment and Simultaneous Tree Estimation

• From the Developers’ webpage (University of Kansas: Jiaye Yu, Mark Holder, Jeet
Sukumaran, Siavash Mirarab, and Jamie Oaks):

SATé is a software package for inferring a sequence alignment and phylogenetic tree.
The iterative algorithm involves repeated alignment and tree searching operations. The
original data set is divided into smaller subproblems by a tree-based decomposition.
These sub-problems are aligned and further merged for phylogenetic tree inference.

Currently, the following tools are supported, and are bundled with the SATé distribution:

ClustalW 2.0.12 (sequence alignment program)

MAFFT 6.717 (sequence alignment program)
MUSCLE 3.7 (sequence alignment program)
OPAL 1.0.3 (sequence alignment program)

PRANK 100311 (phylogeny-aware alignment program)

RAxML 7.2.6 (phylogeny estimator program)

FastTree 2.1.4 (phylogeny estimator program)
!
49
 SATe-2 needs Python 2.7 (Upgrade Python Instructions)
• MAC OS: Open terminal (go the HD>Applications>Utilities>Terminal)

• MAC OS: Check your version of Python

python --version!
!
• MAC OS: if necessary upgrade python to 2.7 as required by SATe-II
!
[Link]

50
 Install SATe-2
• Download SATe-II precompiled from UT-Austin website:

[Link]

***For those more adventurous you can download the command based 'SATe-II' from:

[Link]

download: [Link]

Follow the instructions in the main webpage

51
 Preparing FASTA filed for SATe-2
• Download the FASTA files from the course website

Liolaemus_CYTB.fasta
Lizard_RAG1.fasta

• Create two output folders for the alignment results in your desktop and place the fasta
files in the corresponding one

folder: Liolaemus_CYTB
folder: Lizards_RAG1

52
 Running SATe-2
• Open SATe-II GUI by clicking on the executable on the program folder:

• Explore the console and the options in the SATe-II GUI version:

53
 Running SATe-2
• Explore the options in the SATe-II GUI version:

External Tools:
Aligner: [ClustalW2, MAFFT, PRANK, OPAL]
Merger: [MUSCLE, OPAL]
Tree Estimator: [RAXML, FASTTREE]
Model: [RAxML-options: GTRCAT, GTRGAMMA, GTRGAMMAI;
FASTTREE-options: GTR+G20, GTR+CAT, JC+G20, JC+CAT ]

Sequences and Tree:

Sequence file ...: [This is the folder where our fasta file resides]
Multi-locus Data [option]
Data Type: [DNA, RNA, Protein]
Initial Aligment [option]
Tree file (optional): [Provide if you have an initial phylogeny associated with the sequences]

Workflow Settings:
Algorithm [option] Two-Phase (not SATe-II)
Post-Processing [option] Extra RAxML Search

54
 Running SATe-2
• Explore the options in the SATe-II GUI version:

Job Settings:
Job Name: [give a name for the job]
Output Dir.: [Select the corresponding directory for the output aligment]
CPU(s) Available: [It will depend on your computer]
Max. Memory (MB): [It will depend on your computer]

SATe-II Settings
Quick Set: [Presets: SATe-II_fast, SATe-II_ML, SATe-II_simple, custom]
Max. Subproblem:
Percentage [default 50]
Size [default 10]
Decomposition:
Centroid (fast) or Longest (slow)
Apply Stop Rule: [options]
Stopping Rule: Blind Mode Enabled
Time Limit (hr) [default 24 hours]
Iteration limit [default 1 iterations]
Return: [Default are Final or Best alignment]

55
 Running SATe-2: Select the Following Options
External Tools:
Aligner: [MAFFT]
Merger: [MUSCLE]
Tree Estimator: [RAXML]
Model: [GTRGAMMAI]

Sequences and Tree:

Sequence file ...: [Liolaemus_CYTB.fasta]
Data Type: [DNA]
Tree file (optional): [None]

Workflow Settings:
Algorithm: [None] Two-Phase (not SATe-II)
Post-Processing: [None] Extra RAxML Search

Job Settings:
Job Name: [Liolaemus_CYTB_alignment]
Output Dir.: [Liolaemus_CYTB] Select the corresponding directory for the output alignment
CPU(s) Available: [2] It will depend on your computer
Max. Memory (MB): [1000] It will depend on your computer

SATe-II Settings
Quick Set: [SATe-II_fast]
Iteration Limit: [3]
Leave other options unchanged
56
Running SATe-2

57
 Running SATe-2

• Explore the output in a text editor. The alignment is located in these .aln files in fasta
format:

satejob.marker001.Liolaemus_CYTB.aln

Repeat the same process with

the Lizard_RAG1.fasta file

58
 Mesquite: Visually explore the alignments

• Download mesquite:

[Link]

[Link]

59