gschofl / biofiles Goto Github PK

I have updated annotation data for my Genbank file. Is it possible to create new feature objects and extend the list of features of the given Genbank file ?

Hello, I have an R data frame like the following and I was wondering if it is possible to create a GenBank file out of each one of the entries.

This is an example data frame:

              Barcode     TRAV   TRAJ               TRA_CDR3                 TRA_Full
12 AACxxxxxxxxxxxAT-1   TRAV21 TRAJ41 TGTGCxxxxxxxxxxxxxxTTT ACTxxxxxxxxxxxxxxGGAA...
13 AACxxxxxxxxxxxTC-1   TRAV21 TRAJ41 TGTGCxxxxxxxxxxxxxxTTT GATxxxxxxxxxxxxxxACAA...
27 AAGxxxxxxxxxxxTA-1   TRAV21 TRAJ41 AAAGGxxxxxxxxxxxxxxTTT TGGxxxxxxxxxxxxxxGAGC...
30 AAGxxxxxxxxxxxTT-1   TRAV11 TRAJ31 TGTGCxxxxxxxxxxxxxxTTT ACTxxxxxxxxxxxxxxGGAA...
37 AATxxxxxxxxxxxGG-1   TRAV11 TRAJ31 GCTACxxxxxxxxxxxxxxTTT GTGxxxxxxxxxxxxxxCTCC...
39 AATxxxxxxxxxxxTC-1   TRAV11 TRAJ31 GTGAGxxxxxxxxxxxxxxTTC CCGxxxxxxxxxxxxxxCAGT...

My goal is to make a different GenBank file for each Barcode, including the TRAV and TRAJ info (maybe in SOURCE or some other field), the CDR3 location in the FEATURE field, and finally TRA_Full as the the ORIGIN sequence.

I was wondering whether biofiles would allow me to not have to do this manually. Looking at the write.GenBank function, it accepts >gbRecord instance. It seems like I can create such instance with the gbRecord function, that accepts A vector of paths to GenBank/Embl format records, an efetch object containing GenBank record(s), or a textConnection to a character vector that can be parsed as a Genbank or Embl record. I am hopping the later option can serve my goal.

I think the question here would be how to format my data frame as a textConnection to a character vector that can be parsed as a Genbank or Embl record, so I can eventually print out as GenBank files... is that even possible? I cannot find any example for this particular task.

Many thanks!
Daniel

Hi,
I got an error,What should I do

> gb_file <- reutils::efetch("CP000148", "nuccore", rettype = "gbwithparts", retmode = "text")
[2018-12-06 12:05:43] [error] handle_read_frame error: websocketpp.transport:7 (End of File)
> gb_file
Object of class ‘efetch’ 
LOCUS       CP000148             3997420 bp    DNA     circular BCT 28-JAN-2014
DEFINITION  Geobacter metallireducens GS-15, complete genome.
ACCESSION   CP000148 AAAS03000000 AAAS03000001 AAAS03000002 AAAS03000003
            AAAS03000004 AAAS03000005 AAAS03000006 AAAS03000007 AAAS03000008
            AAAS03000009 AAAS03000010 AAAS03000011
VERSION     CP000148.1
DBLINK      BioProject: PRJNA177
            BioSample: SAMN02598399
KEYWORDS    .
SOURCE      Geobacter metallireducens GS-15
  ORGANISM  Geobacter metallireducens GS-15
            Bacteria; Proteobacteria; Deltaproteobacteria; Desulfuromonadales;
...
EFetch query using the ‘nuccore’ database.
Query url: ‘https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?=efetch&db=nuccore&id=CP000148&retmode=text&rettype=...’
Retrieval type: ‘gbwithparts’, retrieval mode: ‘text’

> rec <- biofiles::gbRecord(gb_file)
Error in { : task 1 failed - "'mc.cores' must be >= 1"
> rec
Error: object 'rec' not found

I'm having this error when I try to write a genbank file:

write.GenBank(gb, 'genome1_g1.gbk')
Error in strsplit(seqs, "(?< = .{10})(? = .)", perl = TRUE) : 
  invalid split pattern '(?< = .{10})(? = .)' 
In addition: Warning message:
In strsplit(seqs, "(?< = .{10})(? = .)", perl = TRUE) :
  PCRE pattern compilation error 
	'unrecognized character after (?<'
	a ' = .{10})(? = .)'

Dear biofiles developpers / users,

I am trying to extract 16S operon from a collection of genbank flat files using R.

I am able to load the file sucessfully:

gbk <- biofiles::gbRecord("~/Projects/ETH/benoit/Bacteroides/200824/GCF_903181445.1_NZ_Bacteroides_fragilis_8E3_BL_hyb_genomic.gbff", 
                          progress = TRUE)

But when it comes to use the biofiles::filter function I can't nake it work:
biofiles::filter(gbk, Feature = "rRNA", product = "16S ribosomal RNA") -> rRNA

biofiles::summary(rRNA)

[[NZ_CAEUHN010000001]]
1857785 bp: Bacteroides fragilis isolate NZ_Bacteroides_fragilis_8E3_BL_hyb, whole genome shotgun sequence.
Error in if (is.atomic(x) || len == 1L && length(nm) <= 1L) { :
missing value where TRUE/FALSE needed

biofiles::getSequence(rRNA)

A DNAStringSet instance of length 24
width seq names
[1] 1857785 AAGTTCTGATAGAACTTAGAAGAGAATGCTCTTTTTACTATTGATTTTAATACTTTTCTCT...AATGACCGTCAATAAATTTTCGACATCCTGAACAGAGCTAATATTGTCCCTTATTGGGAT NZ_CAEUHN010000001
[2] 47339 GTCCGTTTTACCACTATAAATAGTTTCGGAAATACTTACGGTTTGAATGAGAAAAGATGTC...GAACGGATAATAAATTGGATATATTCATTTGTTTTCCAAATAGTTACTACTAAAATGCCT NZ_CAEUHN010000002
[3] 44645 AGAAACATTGATTATCAATGTTCTACAGGATAAACACCCAACTTTACGTCCAAACTGTAAA...GATACCTATTATTGGGCTACAAACATATACGTTATGTAGAATTTATAGAAAAAATAGGGG NZ_CAEUHN010000003
[4] 43518 ACATTCGTTCGTTCTCAACTTCTAAAAATGTTTCGTAAAATTTGACGGTTTGAAAGAGAAA...CAAAGCTTTTGAAGATAGAAATCATACTTTTTAAAGGTATTGATTTTCAGAAGGTTTATC NZ_CAEUHN010000004
[5] 5021 AATTATTGGATACAATTTCCAGAAAGAATAATTAGTTTGCTATTGGAAGATAAATATAAAA...GTACCGACCTCGACAACACTTGCTCCGATGACTGTTTCACCGGTAGCATCCGTAACCACA NZ_CAEUHN010000005
... ... ...
[20] 128200 TAATATTAAAGTGATATTAAAGACTGACTCTAAGGCACTTGATGAAGTGGTAGTAGTAGCT...TGTGTCAAAACGTTGGCACAACCTCCTTTTATATTTTACAGTTCTGCTATATTTTCTTTT NZ_CAEUHN010000020
[21] 109732 AAGTTCTGATAGGACTTGGCATTTCTGCCGGCCTACTCTCTCCGAACCATGTGTTCGCTAC...GCTGTTACTACGACTTCGTCAAGTGCCTTAGCATCAGTCTTTAATATCACTTTGATTATC NZ_CAEUHN010000021
[22] 88286 TCGCCTACCGTCCCGATAGAACTTAGTAAACAGTTTTAAAAACACATATAAACATCTTTAT...GGGAGAATCTTGAAGTGTAAGGATCTTGTTATTAGTTATTTATCTTAAGATATAGGTGTC NZ_CAEUHN010000022
[23] 80205 GTTTGTGTAACTATTGTATCCAACAGTAGCTGCTACCGTAAAGTCACCCTTTTTAGGAGTA...ATGATTGTTTGTCAGAGAAGGACTGCCAAAATGACTGATGACATTGGAAAAACAGGCGCT NZ_CAEUHN010000023
[24] 59327 AGATTCATTTATTTTATCTATATTTGCAAATGAGTATTTATATAAACGTTTAAAGCAAATA...CCTCTTACAGAGGCGTTATACGATAACGTAATAATAAAGATTAGGAAAAGACTTTTCTCA NZ_CAEUHN010000024

The full contigs are exported.
Any idea where I am wrong?

Thanks a ton

You probably are aware of this, but biofiles has been removed from CRAN. I was still able to install it using devtools::install_github, but I hope you'll fix the issues and put it back on CRAN! I found it to work better than genbankr for the GB files I had to work with.

Thanks!

That goes for your reutils package as well - I was about to begin working on this as well as this functionality was missing in Bioconductor, but then I saw your repos...

After building the package with RStudio I get the following error:

installing to Z:/Documents/R/win-library/biofiles/libs/x64
** R
** inst
** preparing package for lazy loading
Error : object 'unlist' is not exported by 'namespace:Biostrings'
ERROR: lazy loading failed for package 'biofiles'

Checking the NAMESPACE file for the recently installed Biostrings packages reveals that unlist is exported but as a method. Testing the package (devtools::test()) produces three additional warnings:

Warnmeldungen:
1: object 'unlist' is not exported by 'namespace:Biostrings' 
2: object 'queryHits' is not exported by 'namespace:IRanges' 
3: object 'split' is not exported by 'namespace:IRanges' 
4: object 'subjectHits' is not exported by 'namespace:IRanges'

I'm using R 3.3.0 64 bit and Bioconductor 3.3 with Biostrings 2.40.2. How to solve this problem?

PS: Installing boost_regex was a nightmare.

Error upon install in Ubuntu 14.04, R 3.3.1, Bioconductor 3.3:

Error in dyn.load(file, DLLpath = DLLpath, ...) : 
  unable to load shared object '/home/rc16041/Software/Rpackages/biofiles/libs/biofiles.so':
  /home/rc16041/Software/Rpackages/biofiles/libs/biofiles.so: undefined symbol: _ZNSt20regex_token_iteratorIN9__gnu_cxx17__normal_iteratorIPKcSsEEcSt12regex_traitsIcEEneERKS7

Files without REFERENCE field raise
Error in { : task 1 failed - "'to' cannot be NA, NaN or infinite"
doe it necessary field?