Hi,

I'm having problems following the biomaRt vignette. I think it comes down to instable servers and if this is the wrong place to report this, please point me in the right direction.

Here is how far I get:

ensembl<-useMart("ensembl")
listDatasets(ensembl, verbose=T)
ensembl = useDataset("hsapiens_gene_ensembl",mart=ensembl)

The error claims:

Error in checkDataset(dataset = dataset, mart = mart) :
  The given dataset:  hsapiens_gene_ensembl , is not valid.  Correct dataset names can be obtained with the listDatasets() function.

Taking one step back and looking at the output of listDatasets, the number of entries in this list fluctuates between 6 and 56. In the vein of this, I found that if I repeat the useDataset it will sometime execute correctly. So even if it can be claimed to work, I find this highly annoying. Is it intentional?

sessionInfo()
R version 3.4.0 (2017-04-21)
Platform: x86_64-pc-linux-gnu (64-bit)
Running under: Debian GNU/Linux 7 (wheezy)

Matrix products: default
BLAS: /DEEP_fhgfs/projects/karln/R/R-3.4.0/lib/libRblas.so
LAPACK: /DEEP_fhgfs/projects/karln/R/R-3.4.0/lib/libRlapack.so

locale:
 [1] LC_CTYPE=en_US       LC_NUMERIC=C         LC_TIME=en_US        LC_COLLATE=C        
 [5] LC_MONETARY=en_US    LC_MESSAGES=en_US    LC_PAPER=en_US       LC_NAME=C           
 [9] LC_ADDRESS=C         LC_TELEPHONE=C       LC_MEASUREMENT=en_US LC_IDENTIFICATION=C

attached base packages:
[1] grid      stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
[1] circlize_0.4.3        biomaRt_2.32.1        ComplexHeatmap_1.14.0 ggplot2_2.2.1        

loaded via a namespace (and not attached):
 [1] Rcpp_0.12.16         pillar_1.2.1         compiler_3.4.0       DEoptimR_1.0-8      
 [5] RColorBrewer_1.1-2   plyr_1.8.4           tools_3.4.0          bitops_1.0-6        
 [9] viridis_0.5.1        class_7.3-14         prabclus_2.2-6       digest_0.6.15       
[13] bit_1.1-12           dendextend_1.7.0     mclust_5.4           memoise_1.1.0       
[17] RSQLite_2.1.0        tibble_1.4.2         gtable_0.2.0         lattice_0.20-35     
[21] viridisLite_0.3.0    rlang_0.2.0          DBI_0.8              parallel_3.4.0      
[25] mvtnorm_1.0-7        gridExtra_2.3        trimcluster_0.1-2    cluster_2.0.7-1     
[29] IRanges_2.10.5       S4Vectors_0.14.7     GlobalOptions_0.0.13 fpc_2.1-11          
[33] bit64_0.9-7          stats4_3.4.0         diptest_0.75-7       nnet_7.3-12         
[37] Biobase_2.36.2       robustbase_0.92-8    GetoptLong_0.1.6     flexmix_2.3-14      
[41] AnnotationDbi_1.38.2 XML_3.98-1.10        blob_1.1.1           kernlab_0.9-25      
[45] magrittr_1.5         whisker_0.3-2        BiocGenerics_0.22.1  scales_0.5.0        
[49] modeltools_0.2-21    MASS_7.3-49          shape_1.4.4          colorspace_1.3-2    
[53] RCurl_1.95-4.10      lazyeval_0.2.1       munsell_0.4.3        rjson_0.2.15 

Sounds like you're experiencing the issue described in biomaRt: drerio_gene_ensembl dataset missing

This was caused by the Ensembl release at that time including some species with apostrophes in their name for the first time, which broke listDatasets().

I suggest updating your version of biomaRt to at least 2.34.1 where a patch was introduced.  You can do this via:

BiocInstaller::biocLite('biomaRt')