Link to this post | posted 01 Apr, 2022 19:45

debbie

Hi Shima, Each time Glimmer and GeneMark predict coding potential, they use a SAMPLE of the genome to create their profiles. So the output that is then displayed in Phamerator and DNA Master can be different because data was generated independently. In my world, that is a good thing. The math of Glimmer and GeneMark breaks down for small open reading frames, and requires personal attention. (It is why we do what we do.) The outputs are showing you that a human is going to have to discern where the gene is and support that call with the data in total. It is also why I use the STOP codon of an open reading frame to identify a gene, because all others numbers can be different. debbie

Link to this post | posted 31 Mar, 2022 15:00
debbie	Exactly!

Link to this post | posted 30 Mar, 2022 19:08
debbie	Hi Pam, In some of our older protocols at Pitt we have some plaque picture taking protocols found here: https://phagesdb.org/workflow/Toolbox/. We use a gel doc to photograph plates. debbie

Link to this post | posted 30 Mar, 2022 17:50
debbie	Hi Ann, No, I don't believe that sequence is a tRNA in the phage, but if you integrate it into smeg (you will find that it completes a tRNA gene in M. smegmatis (tRNA Thr (Msmeg_6152)). It is picked up by tRNAScanSE because it is likely half of a tRNA that is made when the phage is integrated as a prophage. There is nothing to call in the phage annotation for this. Make sense? debbie

Link to this post | posted 29 Mar, 2022 20:24

debbie

Hi Allison, There are a lot of routes that our phages take to integrate, and close to the integrase is a good way to address it. We - I mean Graham, has spent a lot of time looking for integration sites in the Mycobacteria and proving them biochemically. He is writing a review article that gives a great summary, so when it is available it will help a lot. In the meantime, the first thing you want to know is what kind of integrase you have. The large serine integrases can have an attachment site that is but 3 bases in length. so without some bench work, that is going to be difficult. The tyrosine integrases have arm extensions that make it more readily identifiable. You can look up Bxb1 integration (first author is Pallavi Ghosh). Some of our best data published is in M. abscessus (Dedrick, 2021) because we found the prophages and could see the sequences of the attL and attR in the hosts. Those reads should help to get you started. Some of our phages like to integrate into tRNAs. Being able to match the attP to an attB is also needed, so know in the host sequence will be needed. After you read the Dedrick papers, I think you will see why the answer to your questions are not simple. Having said all that, another paper in the works is a prophage finder, that will help identify attL and attR. So much to do!

Link to this post | posted 29 Mar, 2022 12:37
debbie	Kathleen, Yep. It is rule based on a small data set from some mass spec data. Choose the second one. debbie

Link to this post | posted 25 Mar, 2022 12:33
debbie	I agree with Steve. This may be one of the only places where I rely on that particluar pFam hit. debbie

Link to this post | posted 20 Mar, 2022 14:31
debbie	Hi Sally, I think I can call a toxin or antitoxin without its partner because the host may carry it. Curious if others agree. debbie Edited 20 Mar, 2022 14:31

Link to this post | posted 15 Mar, 2022 13:23
debbie	Vicky, Yes, the hierarchy of what to believe is convoluted. We got tail assembly chaperone calls wrong in this cluster. So Blast data - whether at ncbi or phagesdb - will just regurgitate what we got wrong. BUT, if we got it correct, it would be a good source of info. We are steadily working at correcting the 'wrongs', but it just takes time. debbie

Link to this post | posted 15 Mar, 2022 13:16
debbie	Shima, What you need to provide is one best explanation for each gene call. Since i don't use PECAAN this way, I can't answer your question. You are correct, Claire set up PECAAN so there is an output that works. But whatever you pick, be sure to identify the problem areas and the thinking process. debbie