How to subset a set of EST sequences in fasta file based on sequence ID or Name using R?

Question

[second_exon]

I have thousands of EST sequences in a fasta file. How to subset a set of sequences based on sequence ID or name using R?

Sequence example:

>gi|296783888|gb|GW992815.1|GW992815 UAS-Mi10 Complementary DNA of mulberry (Morus indica) Morus indica cDNA 5' similar to Putative phosphoribosyltransferase/phosphoribosylanthranilate-like gene from Morus indica, mRNA sequence
GCAGCCGTCGGATCGTGAGCGTGATCGCGTGGCTAGTCGGGTTGGCGAAATGGTTGGATGATATCCGGAG
GTGGAGGAACCCCATTACCACGGTATTGGTCCACATCTTATATTTAGTGCTTGTTTGGTACCCGGATTTG
ATTGTCCCAACCGGGTTTTTATATGTGTTCCTAATCGGTGTATGGTACTATCGGTTTCGGCCCAAGATAC
CAGCGGGTATGGATACCCGACTCTCACAAGCTGAAGCGGTTGACCCGGATGAGCTTGATGAGGAATTCGA
CACCATACCGAGCTCAAAACCACCCGACATAATCAGGGTCCGGTATGACCGGTTGCGGATATTGGCAGCC
CGGGTTCAAACGGTTTTGGGTGATTTTGCAACACAAGGGGAGCGGGTTCAGGCCTTGGTTAGCTGGAGGG
ACCCAAGGGCCACAAAATTGTTCATAGGCGTGTGCTTGGCCATAACAATAATTCTCTATGTGGTGCCACC
CAAAATGGTTGCCGTGGCACTTGGATTCTACTATTTACGACACCCCATGTTCCGAGACCCCATGCCTCCT
GCAAGCTTGAATTTCTTCAGAAGGCTTCCAAGCCTTTCAGACCGCTTTAATGTAGATTAGAATATTATAT
GATTATTAGTAGGCCCAA

>gi|296783887|gb|GW992814.1|GW992814 UAS-Mi9 Complementary DNA of mulberry (Morus indica) Morus indica cDNA 5' similar to Dehydration-responsive protein RD22, Similar to BURP domain-containing protein like gene from Morus indica, mRNA sequence
AAGCAGTGGTCTAGAACCAGAGTGGCCCCTGCGATGCAGGTATCATCTCTATTATCAAAAGGGATAAGGG
GTGGATCCGTCGGGGATTTGAGTCTCACATGGTCGCTGATAACTTATTGAATGGATATTGGATTGTGTGC
AGTGCGACCTAAACAGGATTGCCGTTGGGGCCTGTGGTCAGAGATACCCCACACTTCTCAACTCCCAAAT
TGGATCTTGTTCCTTGTTTTCCTGTATTAAGCCTGACCCCTGAGGCTTTCGCCACTGCCAACTGGGTGCC
GCCTGCTGACTTCTGATTCCCCGTGCTAACGGTTACTCCCGATTCCTTATCCACATCGAAGATGAACTAT
TGACTTCCGCAAACTCAAAAGGCTGCAAGATATCACTGACCGCTGTCGGGATCCGCGATCGGCATATACG
CGAAATCCGATCCCGGATCCCGGGACTGCAGACGGCTGAA

Like using this header:

>gi|296783888|gb|GW992815.1|GW992815 UAS-Mi10 Complementary DNA of mulberry (Morus indica) Morus indica cDNA 5' similar to Putative phosphoribosyltransferase/phosphoribosylanthranilate-like gene from Morus indica, mRNA sequence

or by using gi number?

How to do this in R?

Answer 1

[Devon Ryan]

library(Biostrings)
f <- readDNAStringSet("sequences.fa")

You can then use the names(f) accessor to either directly match the names of the sequences, grep with gi accession, or even just split the names and directly match the gi accessors. The f object can be directly subset (e.g., f[c(1,3,5,6)]).