Boxplots of field data from 2003. Notice 0 yield in species, F1, F2 and pretty much the BC1F2. The good news was the BC3F1 generation. Good yields, better fiber than cultivated parent.
Monday, February 28, 2005
Saturday, February 26, 2005
Chimaeric clones
Well,
Dr. Fairbanks has pretty much finished with his analysis of my sequences.
Here is the verdict:
14/20 have multiple inserts
That is a lot. That is way more than even I in my worst imaginings thought could go wrong. That means that probably none of my primers are amplifying what we thought they were.
What does that mean for 6 sequences used for phylogenetic analysis? Where they chimaeric?
If yes, then the analysis is bunk.
If no, Then maybe, assuming there isn't any other HUGE messups then the analysis is legitimate.
What does that mean for the minisatellite sequences?
There were thirty something of those. We need to check the area used in the alignment, only 80 bases. Pretty good chance those 80 bases are intact.
Solution: Submit the consensus sequence to Genbank.
What does that mean for the FISH data (Microscopy pictures)?
IF the clones used are not chimaeric then it too is OK, if not then it also needs to be redone.
What does this mean for the microsatellites?
14/20 are probably not reliably amplifying from one region.
That is a huge problem, especially considering the number of short repeat sequences.
Solution
1. So the key with those might be to pick the best of the repeats and go over the sequences to see if they have multiple inserts.
2. Keep the ones that are good and
3. test to see if sequence size is consistent in 'Real'.
4. Test for polymorphism in other entries in panel used by Shanna.
Ultimate solution.
Two words. Mental breakdown
Dr. Fairbanks has pretty much finished with his analysis of my sequences.
Here is the verdict:
14/20 have multiple inserts
That is a lot. That is way more than even I in my worst imaginings thought could go wrong. That means that probably none of my primers are amplifying what we thought they were.
What does that mean for 6 sequences used for phylogenetic analysis? Where they chimaeric?
If yes, then the analysis is bunk.
If no, Then maybe, assuming there isn't any other HUGE messups then the analysis is legitimate.
What does that mean for the minisatellite sequences?
There were thirty something of those. We need to check the area used in the alignment, only 80 bases. Pretty good chance those 80 bases are intact.
Solution: Submit the consensus sequence to Genbank.
What does that mean for the FISH data (Microscopy pictures)?
IF the clones used are not chimaeric then it too is OK, if not then it also needs to be redone.
What does this mean for the microsatellites?
14/20 are probably not reliably amplifying from one region.
That is a huge problem, especially considering the number of short repeat sequences.
Solution
1. So the key with those might be to pick the best of the repeats and go over the sequences to see if they have multiple inserts.
2. Keep the ones that are good and
3. test to see if sequence size is consistent in 'Real'.
4. Test for polymorphism in other entries in panel used by Shanna.
Ultimate solution.
Two words. Mental breakdown
Wednesday, February 16, 2005
Committee review
Odd screen getting to here. Wonder what blogger is changing?
Anyway,
This last few months I have been trying to get my stuff published from my masters. It has been a nightmare. First of all, my committee there is slower than the development of dirt in getting edited drafts back to me. Then, they have all the original data: lab notebooks, databases, etc. But they can't find anything in it. So I have summaries and batch text files. These I try to peice together to answer questions I never got while I was there because they NEVER paid much attention I guess.
This is the thing that bugs me. When I wrote my thesis. These were the comments, "Best thesis I ever read. Minor changes only." "Why did you use these sequences in the alignment and not others?"
Answer: "Because they were the best sequences, ie fewest unkowns, longest, etc. "
Response: "Oh"
No one once questioned the length of my sequencing reads, the quality of library construction, the scoring of the SSR gels, the development of the primers, etc etc. Now, TWO years later, suddenly, there are problems with all of it.
I am willing to admit I did not due it all right, but at the time I presented the exact same data, the same procedures, the same results, the same methods, the same conclusions, and they were the "Best thesis I ever read." Now they are not.
I suspect they didn't even bother to read the original thesis when I wrote it. They just assumed everything was fine, which is dumb, because I could have fixed problems at that point. Now it just makes me look like a bad researcher.
Anyway,
This last few months I have been trying to get my stuff published from my masters. It has been a nightmare. First of all, my committee there is slower than the development of dirt in getting edited drafts back to me. Then, they have all the original data: lab notebooks, databases, etc. But they can't find anything in it. So I have summaries and batch text files. These I try to peice together to answer questions I never got while I was there because they NEVER paid much attention I guess.
This is the thing that bugs me. When I wrote my thesis. These were the comments, "Best thesis I ever read. Minor changes only." "Why did you use these sequences in the alignment and not others?"
Answer: "Because they were the best sequences, ie fewest unkowns, longest, etc. "
Response: "Oh"
No one once questioned the length of my sequencing reads, the quality of library construction, the scoring of the SSR gels, the development of the primers, etc etc. Now, TWO years later, suddenly, there are problems with all of it.
I am willing to admit I did not due it all right, but at the time I presented the exact same data, the same procedures, the same results, the same methods, the same conclusions, and they were the "Best thesis I ever read." Now they are not.
I suspect they didn't even bother to read the original thesis when I wrote it. They just assumed everything was fine, which is dumb, because I could have fixed problems at that point. Now it just makes me look like a bad researcher.
Tuesday, February 08, 2005
One of those days
Have you ever had one of those days where nothing quite works?
Today, Ben, an undergraduate student, started extracting DNA samples and then I finished the protocol. He has successfully extracted DNA samples hundreds of times. So I wasn't worried. I didn't watch what was going on much, just took over when he had to leave at 3:30. It is now 9:30. I have all 115 samples in 95% ethanol.
NO PELLETS = NO DNA . !!!!!!!!!
These samples were irreplaceable. The plants from the field that we phenotyped are gone. Tissue samples are used up and I only have fiber and seed from them buried deep in a pile of other samples to be ginned.
Oh well, I didn't want to do that QTL type study anyway.
Today, Ben, an undergraduate student, started extracting DNA samples and then I finished the protocol. He has successfully extracted DNA samples hundreds of times. So I wasn't worried. I didn't watch what was going on much, just took over when he had to leave at 3:30. It is now 9:30. I have all 115 samples in 95% ethanol.
NO PELLETS = NO DNA . !!!!!!!!!
These samples were irreplaceable. The plants from the field that we phenotyped are gone. Tissue samples are used up and I only have fiber and seed from them buried deep in a pile of other samples to be ginned.
Oh well, I didn't want to do that QTL type study anyway.
Saturday, February 05, 2005
Thursday, February 03, 2005
Seminar
I kinda muffed seminar this week. First I forgot to shave or wear anything nice until I was sitting at my desk with no time to do anything about it. I was even wearing my hawaiian shirt with small holes from splashes of sulfuric acid from delinting seeds. That's what happens when I get dressed before I really am awake I guess.
Then i was nervous and Stelly spent his introduction selling the beauty and wonder that is my experiment. Which involves wild species in a generation mean analysis where we can't analyze it as a generation means analysis due to low sampling in the F2 generations. AAAAAAaaaaaaah. Then I was too negative and didn't emphasize the positive, which was we had improved fiber qualities in both experiments, even if there was a yield drop. And the must BC3F1s had as good of yield with improved fiber across the board.
Since only Stelly from my committee was there I guess that it is OK.
Repeat after me,
AAAAAAAAAAAAAAAAAAAAAaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaahhhhhhhhhhhhhhhhhhhhhhhhhhhhh
Then i was nervous and Stelly spent his introduction selling the beauty and wonder that is my experiment. Which involves wild species in a generation mean analysis where we can't analyze it as a generation means analysis due to low sampling in the F2 generations. AAAAAAaaaaaaah. Then I was too negative and didn't emphasize the positive, which was we had improved fiber qualities in both experiments, even if there was a yield drop. And the must BC3F1s had as good of yield with improved fiber across the board.
Since only Stelly from my committee was there I guess that it is OK.
Repeat after me,
AAAAAAAAAAAAAAAAAAAAAaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaahhhhhhhhhhhhhhhhhhhhhhhhhhhhh
Subscribe to:
Posts (Atom)