Determine Amplification Bias - ei

[weshorton]

Summary

In PCR1, we use a multiplex reaction of 20 forward and 13 reverse primers. Each primer has a different amplification rate during the reaction, and we need to characterize each amplification rate.

• Example
  • Clonotype A is amplified by V1J1 primer combination, Clonotype B is amplified by V2J2 combination.
  • Sample contains equal starting amounts of Clonotype A and Clonotype B
  • Post-amplification/sequencing, sample contains twice as many counts of Clonotype A than Clonotype B
  • V1J1 has double the efficiency of V2J2

    Significance

We can't know the initial counts of clonotypes within a sample, and can only measure them post-amplification and sequencing. We need to incorporate this into our model (Wiki - Model) in order to determine the correct count for each clonotype.

To Do

1. Determine primer independence (link to issue)
2. Depending on above, calculate 260 ei factors, or 33

   Approach

3. Control Experiment
   • Samples containing:
     • Constant spike concentration (To Do: what is this value?)
     • NO gDNA
   • 20 total samples
4. Data Analysis and Interpretation
   • Boxplots of spike counts
     • for all 260 primer combinations
     • grouped by V primer (20)
     • grouped by J primer (13)
   • ANOVA?
     • predictor variable - primer (or primer combination)
     • response variable - count
     • Ho: No variation between counts for different primers
     • Ha: Spike count deviates significantly from mean count depending on primer used.
   • Difference from mean used to calculate ei factor
     • How?
5. Relevant plots, tables, and other information
   • [Box plots] of spike counts vs primer identity
   • [Table] of ANOVA results
   • [Table] of ei factors

[weshorton]

Potential cross-amplification between V13-1, V13-2, and V13-3. Compare sequence similarity to see how possible this is.

V13_Blasts.txt

V13-1 to V13-2 - 88% V13-1 to V13-3 - 83% V13-2 to V13-3 - 89%

Did not have significant matches to other V genes.

[weshorton]

To Do

1. Turn into milestone
2. How is amplification bias captured in our new normalization method?
   • Talk with Burcu
     1. If possible, determine general amplification bias. If not, must determine for each sample/batch.