Protocols:HeliScopeCAGE: Difference between revisions
From FANTOM5_SSTAR
No edit summary |
|||
| Line 14: | Line 14: | ||
* 5% error rate is on these HCO control channels is 'requirements' for standard sequencing (while it sometime shows ~5.5% or such) | * 5% error rate is on these HCO control channels is 'requirements' for standard sequencing (while it sometime shows ~5.5% or such) | ||
* Channel 1, 13, and 25 are not very stable, in comparison with other channels (error rate can be much higher, sometimes). | * Channel 1, 13, and 25 are not very stable, in comparison with other channels (error rate can be much higher, sometimes). | ||
* Sequencing starts by fill and lock - you can't see the 1st base (or first non-T, if it starts with T) | * Sequencing starts by fill and lock - you can't see the 1st base (or first non-T, if it starts with T). | ||
* The sequencing length are variable even in one channel (mainly depending on base order of the reads). Mean length is around ~33nt ('filtered' reads are typically from 20nt - 50nt) | * The sequencing length are variable even in one channel (mainly depending on base order of the reads). Mean length is around ~33nt ('filtered' reads are typically from 20nt - 50nt) | ||
* Helicos software to process raw reads is HeliSphere http://open.helicosbio.com/helisphere_user_guide/ | * Helicos software to process raw reads is HeliSphere http://open.helicosbio.com/helisphere_user_guide/ | ||
* filterSMS is the tool to discard artifacts and unusable reads. Practically, it discards (i) too short sequence (<20nt), (ii) CTAG repeats (termed BAO, Base Order Addition) | * filterSMS is the tool to discard artifacts and unusable reads. Practically, it discards (i) too short sequence (<20nt), (ii) CTAG repeats (termed BAO, Base Order Addition) | ||
* Many of the 5% errors are on insertion/deletions. Accordingly, we need (full/semi) dynamic programing based algorithm to align reads. For that purpose we have developed delve alignment program (Timo Lassmann) | * Many of the 5% errors are on insertion/deletions. Accordingly, we need (full/semi) dynamic programing based algorithm to align reads. For that purpose we have developed delve alignment program (Timo Lassmann) | ||
Revision as of 15:51, 30 August 2012
We started HeliScope CAGE with manual operation (with multi(8)-channel pipet), and achieved also automated preparation of the CAGE libraries to scale up the number of profiles (96 RNAs can be treated at once).
HeliScopeCAGE library construction
- Manual library preparation (standard): OP-HELISCOPE-CAGE-v3.12
- Automated library preparation: OP-HELISCOPE-CAGE-v4.0
- HeliScope sequencing: OP-HELISCOPE-sequencing-v1.0
- Filtering with HeliSphere/filterSMS: OP-HELICOS-CAGE-Filtering-v1.0
Known facts about HeliScope technology
- 1 sequencer has two flow cells, each of them has 25 channels (corresponding to 'lanes' - in Illumina)
- Channel 25 on both flow cells are dedicated to sequence Helicos Control Oligo (termed HCO), which is the artifactual oligonucleotide which we already know the sequences. The performance of the sequencers/runs are mainly checked by this sequencing.
- 5% error rate is on these HCO control channels is 'requirements' for standard sequencing (while it sometime shows ~5.5% or such)
- Channel 1, 13, and 25 are not very stable, in comparison with other channels (error rate can be much higher, sometimes).
- Sequencing starts by fill and lock - you can't see the 1st base (or first non-T, if it starts with T).
- The sequencing length are variable even in one channel (mainly depending on base order of the reads). Mean length is around ~33nt ('filtered' reads are typically from 20nt - 50nt)
- Helicos software to process raw reads is HeliSphere http://open.helicosbio.com/helisphere_user_guide/
- filterSMS is the tool to discard artifacts and unusable reads. Practically, it discards (i) too short sequence (<20nt), (ii) CTAG repeats (termed BAO, Base Order Addition)
- Many of the 5% errors are on insertion/deletions. Accordingly, we need (full/semi) dynamic programing based algorithm to align reads. For that purpose we have developed delve alignment program (Timo Lassmann)
