Frequently Asked Questions
| • | If the TruSight Tumor 170 Panel is selected along with a custom manifest, which file is used? |
| • | The TruSight Tumor 170 manifest is used and overrides the custom manifest input. |
NOTE
FASTQ files may need to be renamed for use in local analysis of UMI Error Correction Local App, due to the Local App expecting the FASTQ files to be named using the Sample_ID. BaseSpace Sequence Hub also converts any underscores to dashes in the output FASTQ Files.
| • | How many samples can a specific instrument multiplex using UMIs? |
| • | Recommended rule is for every 100 kb of panel, 50–60 million PF clusters are needed. |
| • | HiSeq 2500 |
| • | 2B clusters / (60M * (PanelSizeinMB/100Kb)) = Multiplex Samples |
| • | HiSeq 4000 |
| • | 2.5B clusters / (60M * (PanelSizeinMB/100Kb)) = Multiplex Samples |
Example: (1Mb panel): 2.5B clusters / (60M * (1MB/100Kb)) = 4 Samples
| • | What might happen if the amount of DNA is too high or low or has poor fragmentation? |
| • | If the DNA input is too high, the MTC will be very high, leading to small family size and poor error correction. This may cause an ‘outside of threshold’ result for the mean family size >= 10 metric. |
| • | If the DNA input is too low, the MTC will be very low, leading to over sequencing (ie large family size) and may cause an ‘outside of threshold’ result for the MTC >= 954 metric. Over sequencing may lead to increased inherent error rate. |
| • | For cfDNA applications, no fragmentation is needed (they are naturally fragmented). For genomic DNA samples, it is recommended that you fragment to median size ~160 bp to mimic cfDNA size range. |
| • | Interpretation of results: What conclusions can be drawn from the following scenarios? Is the data usable? What can I do to correct any observed issues? |
|
Meets Median Target Coverage Threshold |
Meets Mean Family Depth Threshold |
Meets Noise Allele Frequency Threshold |
Conclusion |
|---|---|---|---|
|
Yes |
Yes |
Yes |
Data are usable for ultralow AF detection (~0.4%). |
|
Yes |
Yes |
No |
Unlikely scenario. If observed, the sample may have significant DNA damage before library prep. |
|
Yes |
No |
Yes |
Slightly insufficient sequencing however, data might still be good to use for variant detection. |
|
Yes |
No |
No |
Insufficient sequencing for DNA input amount. Family size is too small to effectively reduce error rate. |
|
No |
No |
Yes |
Low DNA input or problem in library conversion efficiency. Slightly under sequencing. Not recommended for ultralow AF detection due to low MTC. |
|
No |
Yes |
Yes |
Low DNA input or problem in library conversion efficiency. Sufficient sequencing. Not recommended for ultralow AF detection due to low MTC. |
|
No |
Yes |
No |
Low DNA input or problem in library conversion efficiency. Sufficient sequencing but sample may have significant DNA damage before library prep. |
|
No |
No |
No |
Low DNA input or problem in library conversion efficiency. Under sequencing. Family size is too small to effectively reduce error rate. |
NOTE
Median Target Coverage Threshold is dependent on panel.
Mean Family Depth and Noise Allele Frequency Thresholds are independent of panel but are source material dependent. Sheared gDNA inputs may not meet thresholds.