454-Titanium Sample Submission

Including enhancements from 454 manufacturer (Roche) recommendations.


1. Sample requirements for 454-Titanium sequencing:

Sample requirements depend on the starting materials:
  • Genomic dsDNA ready for nebulization [>5 µg in a 10 µL TE].
  • dsDNA fragments 200 - 700 bp [>3 µg in 10 µL TE].
  • PCR products that incorporate fusion primers must be <500 bp doublestrand [~100 ng in 10 µL TE].

2. Quality of DNA required:

A spectroscopic measurement of the A260/280 ratio of ~1.8 provides a good estimate of sample quality. We also request a 2% agarose gel image of the DNA sample(s) run with a 1 kb ladder to assess incoming sample quality in comparison to our internal QC.

3. Quantity of DNA required:

Sample quantity and quality are determined by agarose gel. You can use fluorometry Agilent bioanalyzer and provide the results. If sample quantity or quality is not adequate, a replacement sample will be needed.

4. Best methods of Quantification:

We prefer agarose gel with mass standards. However, many researchers use fluorometry. Please note that many nano-scale fluorometers can dramatically under-estimate sample quantity which can delay projects if additional an sample must be requested for processing. We recommend using larger sample volumes for quantification.

5. Best buffer to resuspend samples:

Samples should be submitted in a 10 µL volume of 1X TE buffer.

6. What is the accuracy of the sequences that I receive?

The GS-FLX provides a single-read accuracy of 99.5% or better for each individual read of 200-300 bp. This corresponds to roughly a Q40 quality score. This allows consensus accuracy of 99.99% with appropriate depth.

7. What coverage is needed for my genome sequencing project?

20-fold coverage of single-ended reads is recommended for de novo sequencing initiatives which will yield a high quality draft assembly. Incorporating Paired End Reads will aid in assembly and orient the scaffold the resulting contigs.

15-fold coverage is recommended for re-sequencing applications and mapping against a reference sequence to have high confidence in variant calling.

For Ultra-Deep sequencing applications, the depth of coverage is dependent upon application type and experimental goals.

8. How many samples can I analyze per sequencing run?

The number of samples analyzed in parallel depends upon the total size of the samples and the desired depth of coverage. Sequencing can be performed in one, two, four, eight, or sixteen sample formats based on gasket applied to the PicoTiter plate. Additionally 454 allows more flexibility for combining samples for parallel sequencing.

9. What are Multiplex Identifiers (MIDs)?

Roche's MID sequencing adaptors incorporate 10 bp tag sequences that are designed to take into account the instrument nucleotide flow order and ensure that >5 sequencing errors are required to misidentify a read. This allows multiple sequencing libraries to be pooled and the reads sorted by sample during subsequent bioinformatics analysis.

10. How long do projects take?

The length of a project depends on number of samples, genome size and data analysis requirements. The average turn around time for standard projects is 4-6 weeks from project start date, assuming there are no sample quality issues.

11. What is the difference between a draft and a finished genome?

There is no universal consensus on these definitions as is discussed in "What is Finished, and Why Does it Matter" Mardis et. al. Genome Research. Vol. 12, Issue 5, 669-671, May 2002. The definitions below are generalized but may not apply to all circumstances.

A finished sequence defines a highly-accurate genome assembly typically with less than 1 error for every 106 bases and contiguous sequences covering all reliably sequenced regions arranged in the appropriate order. The assembly has typically had gap closure via directed PCR and subsequent sequencing.

A draft genome assembly typically results from computational assembly of shotgun reads and generally has not incorporated biochemical confirmation of the resulting assembly.
The University of Maryland Center for Bioinformatics & Computational Biology has an excellent genome assembly primer covering the complexities of genome assembly.


Contact Robert Bogden for a Quote