Improving the efficiency of immune repertoire exploration

Abstract

In the adaptive immune system T cells are able to recognize a variety of foreign antigens because their T cell receptors (TCRs) appear in diverse structures which are the results of complex V(D)J recombination of TCR genes The recombined TCR gene sequences of all T cells can thus be used to characterize immune repertoire To capture all possible V(D)J recombinations a popular approach is applying multiple primers that target all possible V and/or J regions of TCR genes for amplification This multiplex PCR (mPCR) approach however usually introduces primer bias To avoid primer bias a 5’ rapid amplification of cDNA ends (5’ RACE) approach can be used to ampify TCR genes In the 5’ RACE data however both regularly and non-regularly recombined TCR sequences exist and the later of which could not be used to characterize immune repertoire Current tools may mistake non-regular TCR sequences as regular and report false V(D)J annotations In this thesis we developed a new computational tool TRIg to correctly handle both regular and non-regular TCR sequences in the 5’ RACE data To promote the 5’ RACE approach we further studied the difference between mPCR and 5’ RACE approach We found that 5’ RACE achieved a higher consistency and captured more VJ recombination events than mPCR suggesting less primer bias of the 5’ RACE approach Finally we improved a 5’ RACE method to reduce the proportion of non-regular TCR sequences via carefully controlling size of the PCR amplicons This increased the fraction of useful data from 80% With all these efforts we provide an efficient experimental and computational pipeline for studying immune repertoire

Date of Award	2020
Original language	English
Supervisor	Tsung-lin Liu (Supervisor)