Determining paired immune receptor chains from frequency matched subunits
The invention is directed to methods for determining nucleic acids that encode immune receptor chains originating from the same cell, that is, paired immune receptor chains. Methods of the invention comprise high-throughput sequencing of rearranged nucleic acids encoding immune receptors from one or more samples of lymphocytes. In one aspect, from a plurality of subsets of a sample, nucleic acids encoding separate chains of a pair are separately sequenced, wherein the size of the sample and the number of subsets are selected so that the distribution of lymphocytes approximates a binomial model. Paired chains are determined by identifying pairs that appear together or that are entirely absent in the subsets.
1. A method of determining a profile of likely paired immune receptor chains in a sample containing a plurality of lymphocytes, wherein the sample is obtained from a subject, and wherein the lymphocytes express a repertoire of immune receptor chain pairs, the method comprising the steps of:
(a) partitioning the sample into a plurality of subsets such that each of the subsets comprises at least 1,000 T cells or at least 1,000 B cells;
(b) amplifying VDJ regions and VJ regions of CDR3 sequences of substantially all immune receptor chains in each of the subsets by a single multiplex polymerase chain reaction and sequencing the resulting amplicons to produce a subset list of VDJ regions and VJ regions nucleotide sequences for each of said subsets of step (a);
(c) determining, for each of said subsets, (i) whether or not a first nucleotide sequence encoding a first chain of a likely immune receptor pair in any of the subsets is present in each subset list of nucleotide sequences, and (ii) whether or not a second nucleotide sequence encoding a second chain of a likely immune receptor pair in any of the subsets is present in each subset list of nucleotide sequences; and
(d) producing a profile of a plurality of likely paired immune receptor chains by matching those first and second nucleotide sequences that are present together in at least one subset list and not present separately in other subset lists at all.
2. The method of claim 1 , further comprising the step of repeating said steps (a)-(d) one or more times for another plurality of subsets different from any previous plurality.
3. The method of claim 1 , wherein said immune receptor chains are T cell receptor α chain and T cell receptor β chain.
4. The method of claim 1 , wherein said immune receptor chains are B cell receptor heavy chain variable region and B cell receptor light chain variable region.
5. The method of claim 1 , wherein said sample contains a population of said lymphocytes expressing said pairs of immune receptor chains, and wherein the population has a size and each different lymphocyte of the population has a frequency within the population; and wherein the sample comprises at least 10,000 unique clonotypes.
6. The method of claim 1 , wherein said first immune receptor chain is a TCRγ chain and said second immune receptor chain is a TCRδ chain.
7. The method of claim 1 , wherein said profile of likely paired immune receptor chains comprises at least 100 pairs of nucleotide sequences encoding a first immune receptor chain and a second immune receptor chain and wherein each of said nucleotide sequences has a length in the range of from 30 to 500 nucleotides.
8. The method of claim 7 , wherein said profile comprises at least 1000 pairs of said nucleotide sequences.
9. The method of claim 1 , wherein step (b) generates sequence reads of at least 1000 distinct clonotypes per run.