Discrepancies between sequencing runs

[bshim181]

Hello,

I am analyzing a bulk TCR dataset with UMIs.
From one sample aliquiot, we distributed it evenly into 4 to generate 4 sample replicates.
We have sequenced these 4 replicates twice and compared the outputs from two sequencing runs.
We are noticing significant discrepancies in clonal frequency between two sequencing runs on the same clone.

For example for the clone with,
CAVTDSNYQLIW | TRAV1-2 | TRAJ33

UMI Fraction For Sequencing Run 1

UMI Fraction For Sequencing Run 2

I am wondering what might be causing these discrepancies between replicates and across sequencing runs. I am getting approximately similar alignment and assemble report statistics.

[mizraelson]

It’s hard to say, but clearly something is off during sample preparation. The numbers clearly show a tenfold difference.