Which Length Distribution Represents Higher Quality Run For 454 Data?
3
0
Entering edit mode
12.4 years ago

length distributions

These are two 454 runs for the same source of genomic dna. I am about to decide which set of sequences to use. Therefore, I would like to ask, has one of these runs better quality (based on judging length distribution) or not? Thanks a lot for the answers from more experienced users of 454 data.

454 length • 2.4k views
ADD COMMENT
2
Entering edit mode
12.4 years ago

If these two runs come from the same library, and differ only by the run, then I would consider the first run to be "better" because of the narrower peak around 500bp.

If these are runs from different libraries made from the same genomic data, then I would have a look at the differences in the library protocols (if any -- differences) and in the 454 machine settings, and definitely to the QC report from the 454 machine.

Aside from that, I believe the definition of a "better" run should be based on average read quality distribution rather than on read length distribution.

ADD COMMENT
0
Entering edit mode

Could you please explain to me, why do you consider narrower peak better? (in case everything else is same:) Thanks.

ADD REPLY
1
Entering edit mode

Because a 454 machine runs with just enough reagents to sequence a given length, so if everything else is the same (and you should carefully double check this), I would expect the broader peak to come from slight problems in the 454 reaction process. This is why, I would specifically check if the 454 machine settings were exactly the same.

ADD REPLY
0
Entering edit mode

Thanks for explanation.

ADD REPLY
1
Entering edit mode
12.4 years ago
JC 13k

why not use both?

ADD COMMENT
0
Entering edit mode
12.4 years ago
Vikas Bansal ★ 2.4k

I think one cannot decide about better quality based on lengths alone, if lengths are comparable (I say comparable because if you have 1 graph where most of the sequences are 50-100bp long and 2nd graph shows you, say, 400-500bp then obviously something is wrong). In this case both the graphs (just a note- y axis scale is different) show same distribution. You can download both the files and run QC and then you can decide or you can read the paper (or any other source from where you took the data) that why they run same sample 2 times and Did they change anything during the 2nd run (library prep. etc) and what were there results (in terms of which run is better).

ADD COMMENT

Login before adding your answer.

Traffic: 2051 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6