CHAMPS Antibody Sequencing Workflow

A couple of months ago, we have announced our CHAMPS antibody sequencing service. With the FREE "blind trials" we offered during the promotional period, we have received quite a few dataset from several users. The responses to the results we provided are remarkable.

We require the sample to be reduced with DTT, alkylated with iodoacetamide. Glycans need to be removed and heavy/light chains must be separated. Each chain then will be digested with six enzymes: AspN, chymotrypsin, GluC, LysC, pepsin and trypsin. MS/MS spectra is required to be acquired using LTQ-Orbitrap at high resolution with HCD fragmentation. In total, we require six LCMS runs per chain.

The data analysis starts off from PEAKS de novo sequencing. A list of high quality de novo sequences will be generated along with the positional confidence score for each amino acid. Then an in-house developed program will be used to assemble the de novo peptides into much longer sequences, protein contigs. In our experiments, the majority of assembled contigs had a length of 60~120 residues.

We blast the protein contigs in NCBI nr database to assemble the antibody template. We select a protein hit corresponding to the constant region and select the closest protein hit corresponding to the variable region. All the contigs will then be mapped to the template to get the first draft of the antibody sequence. In principle, we trust contigs in variable region and the template in constant region.

The draft sequence will be refined iteratively using PEAKS SPIDER homology search. During each iteration, we will examine insertion/deletion/mutation reported by SPIDER, residues with low peptide coverage, residues at the protein n-terminus and compare the sequence mass with protein intact mass, if available.


*Some content in this post is extracted from the ASMS 2013 poster "Whole Protein de novo Sequencing from MS/MS". You can find a web version of the poster here.