I need to retrieve the sequences of proteins that I know to be expressed but weren't identified during genome annotation in a given organism (sometimes, entire clades). I know gene, have sequences of homologous proteins and have a HMM of multiple sequence alignment of the same protein in other organisms. Any ideas?
Dear Andre,
So your question is more about the annotation or protein sequence???
In order to combine both, probably, you should have a look on protein annotation pipelines, e.g. from NCBI (https://www.ncbi.nlm.nih.gov/genome/annotation_euk/process/) or Ensembl (https://www.ensembl.org/info/genome/compara/homology_method.html).
The pipelines can be different, depending of the organism reign.
Anyway one of the critical step is to identify functional domains and motifs, because, sometimes, you cannot find directly annotated protein from another specie, even though functional protein regions are identified.
It would give you some hints whether you should search also in distantly related species, but it would depend on the type of the proteins (highly mutated or conservative evolutionary). Dont forget about protein synthesis and functioning localization, as I've met a case when GO (Gene Ontology) annotation claimed 2 opposite cases (e.g. located in mitochondria ONLY; located elsewhere except mitochondria :).
Hope it helps you
Good luck
Dear Levitchi,
Many thanks for your answer.
My interest is in protein sequences only.
My main problem is at retrieving protein sequences that I find in several organisms but fail to get within specific clades. This is not because the protein in question is not expressed in these clades (it is) but simply because the genomes of the organisms in these clades were not annotated properly for some reason by the automated annotation pipelines that you referred (thank you for it btw).
Actually... I may not be finding these proteins in specific clades because they are encoded in mitochondrial genomes (which were not sequenced?) and later migrated to the nuclear genome during evolution in the other clades. I'll investigate this further.
Thank you once again for the input :)
André Rosa
Dear Andre,
Once there is no reference from your side on the organisms you are investigating, I can just speak theoretically. My experince includes the following:
- looking for some bacterial virulent proteins, we realized that some of their genes were not presented in the GeneBank, so, in case, we applied any of the phylogenetic clustering method, there were a lack of branches in order to have reliable results.
- working on sunflower genetics at the time no genome was yet published, we used Arabidopsis as an intermediate reference genome, from which we looked to identify ESTs, generally with poor annotation with the reference to the proteins, but our logic was the combination between Arabidopsis annotation + homology to functional regions of the ESTs as a way to guess such protein exists in Sunflower
- another case was to identify some defensin proteins, without proper annotation as they were under some patents coverage.
So, some of the cases may be fulfilled with biological knowledge for some statements, but as I donot know your goal wouldnot like to make more fantasies ;)
Good luck.
- Badges
- Science topic
- Similar topics
- Entomology
- Resistance