The standard sequencing protocol for cholera recommends the inclusion of at least one negative control containing a water sample. Generally the negative controls are included in the sequencing pool and sequenced, but the data is ignored post-sequencing. Bacpage could allow the user to indicate which samples represent negative controls and clean data (however we define that) and/or warn the user when the negative control contains too much data or when a sample shares a high proportion of sequences with the negative control.
Shoutout to @pvanheus for raising this issue in person.
Options for cleaning the data:
- Generate blast database from negative control, and remove reads from other samples if they are identical and within a similar size (or not to measure pre-tagementation contamination)
- Hash sequences in the negative control, and remove sequences from samples that share the same hash (more conservative)
- Compare samples to negative control, throw out sample if too similar based on some thresholds. Could optionally just warn use about these sequences, by adding a column to the assembly QA report.
The standard sequencing protocol for cholera recommends the inclusion of at least one negative control containing a water sample. Generally the negative controls are included in the sequencing pool and sequenced, but the data is ignored post-sequencing. Bacpage could allow the user to indicate which samples represent negative controls and clean data (however we define that) and/or warn the user when the negative control contains too much data or when a sample shares a high proportion of sequences with the negative control.
Shoutout to @pvanheus for raising this issue in person.
Options for cleaning the data: