基于UniprotID及目标残基批量获取±6位13mer肽段的方法咨询

Hey Emanuele, great question! Here are a few practical ways to batch extract those 13mer peptides (target residue plus 6 amino acids before and after) from your list of Uniprot IDs and target positions. I’ll break down options for Python, R, and no-code online tools to fit your workflow:

The easiest way is to use the Biopython library, which lets you fetch Uniprot sequences directly and manipulate them. Here's a step-by-step example:

1. First, install Biopython if you haven’t already:

   pip install biopython
   

2. Then use this script to process your list:

   from Bio import SwissProt
   
   def extract_13mer(uniprot_id, target_residue, target_pos):
       # Handle isoform IDs by stripping the suffix if needed
       clean_id = uniprot_id.split('_')[0]
       handle = SwissProt.get_sprot_raw(clean_id)
       record = SwissProt.read(handle)
       seq = record.sequence
       
       # Uniprot positions are 1-based; Python strings are 0-based
       start = max(0, target_pos - 7)
       end = min(len(seq), target_pos + 6)
       peptide = seq[start:end]
       
       # Optional: Verify the target residue matches to catch position errors
       if seq[target_pos - 1] != target_residue:
           print(f"Warning: Residue at position {target_pos} in {uniprot_id} is {seq[target_pos-1]}, not {target_residue}")
       
       return peptide
   
   # Replace this with your actual input data
   input_data = [("Q7TQ48_", "S", 442)]
   for item in input_data:
       uniprot_id, res, pos = item
       peptide = extract_13mer(uniprot_id, res, pos)
       print(f"{uniprot_id}: {peptide}")
   

   Pro tip: If you have a large list, add a small delay between requests to avoid hitting Uniprot’s rate limits.

For R users, the UniprotR package simplifies fetching sequences, and basic string manipulation will handle the peptide extraction:

1. Install the required package:

   install.packages("UniprotR")
   

2. Use this script:

   library(UniprotR)
   
   extract_13mer <- function(uniprot_id, target_residue, target_pos) {
       # Clean isoform suffix if present
       clean_id <- strsplit(uniprot_id, "_")[[1]][1]
       # Fetch the protein sequence
       seq_df <- GetProteinSequence(clean_id, format = "fasta")
       seq <- as.character(seq_df$Sequence)
       
       # Calculate bounds (both Uniprot and R use 1-based indexing)
       start <- max(1, target_pos - 6)
       end <- min(nchar(seq), target_pos + 6)
       peptide <- substr(seq, start, end)
       
       # Verify residue match to catch errors
       if (substr(seq, target_pos, target_pos) != target_residue) {
           warning(paste0("Residue mismatch in ", uniprot_id, ": expected ", target_residue, ", got ", substr(seq, target_pos, target_pos)))
       }
       
       return(peptide)
   }
   
   # Replace with your actual input data
   input_data <- data.frame(
       UniprotID = c("Q7TQ48_"),
       TargetResidue = c("S"),
       TargetPos = c(442)
   )
   
   # Process all entries
   input_data$Peptide <- mapply(extract_13mer, input_data$UniprotID, input_data$TargetResidue, input_data$TargetPos)
   print(input_data)
   

If you prefer not to code, here’s a workflow using Uniprot’s built-in tools plus basic spreadsheet functions:

1. Batch fetch sequences:

   • Use Uniprot’s batch retrieval feature to upload all your Uniprot IDs (strip isoform suffixes if needed), then export the results as a CSV/Excel file that includes the full protein sequences.

2. Extract peptides with spreadsheet formulas:

   • In your spreadsheet, add a new column for the peptide. Assuming your sequence is in column B and target position in column C, use this formula (works for Excel/Google Sheets):

     =MID(B2, MAX(1, C2-6), MIN(13, C2+6 - MAX(1, C2-6) + 1))
     

     This automatically handles edge cases where the target position is within the first 6 or last 6 residues of the sequence.

3. Verify residues:

   • Add a column to cross-check the target residue:

     =MID(B2, C2, 1)
     

     Compare this to your target residue column to catch any position mismatches.

内容的提问来源于stack exchange，提问作者Emanuele Loro