When downloading protein family alignment from very large families, I usually chose the filtered options from Pfam (Representative Proteomes or Reference Proteomes) in order to have a manageable number of sequences that sampled the entire tree of life.
With the shutdown of Pfam, I haven't found an alternative for doing that. If I query Uniprot with a Pfam code I get all proteins containing a domain, but none of the filtering options can be useful for doing that. Similarly, I can search for a given domain in Interpro,but the results cannot be filtered by such criteria. Is there any way to do that in any of those databases? Alternatively, is there any tool that would take a very large number of proteins (or their Uniprot codes) and filter by choosing those from representative species on the tree of life (low redundancy, and ideally also prioritizing the most reliable sequences when there is redundancy)?
There are several ways to filter protein sequences from Uniprot or Interpro using databases such as Representative Proteomes or Reference Proteomes. Here are a few options:
Filtering protein sequences from Uniprot or Interpro using databases such as Representative Proteomes or Reference Proteomes is a common practice in bioinformatics research. Here are the general steps to filter protein sequences using these databases:
Here are some additional details for each step:
In summary, filtering protein sequences using Representative Proteomes or Reference Proteomes databases involves aligning the sequences of interest against the database, setting a threshold value to filter the aligned sequences, and extracting the filtered sequences for further analysis.
- Badges
- Science method