Link to this post | posted 31 May, 2018 17:39

delesall

I'd like to request adding this function to the approved list The SPP1 phage helicase loader protein is a well-studies phage protein (see reference below). I have seen good HHPred matches to this protein in a number of phages. In Gordonia phage Rofo, gp 57 shows a couple of HHPred matches with 98% probability; 40-43% coverage and low E-values Structural analysis of Bacillus subtilis SPP1 phage helicase loader protein G39P Bailey, S., Sedelnikova, S.E., Mesa, P., Ayora, S., Waltho, J.P., Ashcroft, A.E., Baron, A.J., Alonso, J.C., Rafferty, J.B. (2003) J.Biol.Chem. 278: 15304-15312

Link to this post | posted 03 Jun, 2021 23:49

debbie

I am adding a response from Chris Shaffer: I did a bit on this. Mostly on the question of nomenclature, not on the validity of the specific gene call. Because of this investigation, I am removing the DnaC-like helicase loader from the approved list. I also ran across this site in poking around the web which is kinda interesting site: https://proteopedia.org/wiki/index.php/DnaC_helicase_loader It says there are two types of helicase loaders: DnaC helicase loader (DnaC) transfers helicase to replication origins. The helicase loaders belong to two classes. The ‘ring breakers’ break the hexameric helicase to allow the DNA to pass through. The ‘ring makers’ assemble the helicase monomers to hexamers around the DNA. Here is a good paper that summarizes these two methods: https://pubmed.ncbi.nlm.nih.gov/12906810/ Unfortunately, there is a complication in that the E coli DnaC is a breaker type and the B subtilis DnaC is a maker type so we have a name crash. This is the kind of thing that makes it really easy to make mis-assignments and/or cause a lot of confusion. If we say "DnaC-like" do we mean the E coli or B sublitis version. So I would argue for this situation, the best solution is to update the terms list to just "helicase loader" and avoid the "DnaC like". I will note in my BLAST searches I found two Bacillus phage proteins that Veronique has published using the term "helicase loader" so she agrees with me.

Link to this post | posted 14 Mar, 2023 19:29

EWisner

Re: Helicase Loader I am working on annotating the phage Culver (CQ1) and I am having trouble confirming the putative function of the gene with the stop codon at 62590 bps. The BLASTp shows a majority of the functions as hypothetical with a few being helicase loader. HHpred shows some matches with products similar to helicase loader, but the matches are not as strong or plentiful to be confident in this identification. Hypothetical protein or Helicase loader?

Link to this post | posted 14 Mar, 2023 19:29

EWisner

Re: Helicase Loader I am working on annotating the phage Culver (CQ1) and I am having trouble confirming the putative function of the gene with the stop codon at 62590 bps. The BLASTp shows a majority of the functions as hypothetical with a few being helicase loader. HHpred shows some matches with products similar to helicase loader, but the matches are not as strong or plentiful to be confident in this identification. Hypothetical protein or Helicase loader?

Link to this post | posted 16 Mar, 2023 21:08

cdshaffer

I would discount the observation that "majority of the functions as hypothetical" because most of those "hypothetical proteins" were probably annotated before the above discussion and thus before the addition of "helicase loader" to the approved terms. So in this case, "absence of evidence is NOT evidence of absence". New crystals are constantly being published and the approved terms list is a living document that changes as we find new functions and fine tune the nomenclature. This is one of those places where I would tell a student "This is why we keep doing annotation by hand, we keep learning more and more and we keep getting better and better at annotation". Looking at the positive evidence though, this call is indeed tricky, there are several HHPRED hits suggestive of helicase loader that all have really high probability but only about 40-50% coverage. So this is where reasonable annotators can disagree. In looking at the crystal data here I can see that the part that does not align is "disordered" so one could use that to argue that a strictly similar structure in this region is not required for function (as this region is not highly structured in the crystalized protein) and thus the fact that it does not match at the structural level is not good evidence that this new protein is not another example of a helicase loader. Bottom line, the fact that this region is disordered means that I discount the evidence that HHPRED is not matching them (i.e. it weakens the negative result). I don't think I ever like adding annotations on just a single piece of weak evidence, even if I can make a handwavy argument for why it is weak, so I would want more evidence. My own sense would be to look for synteny evidence to strengthen the call for a helicase loader. Since proteins that interact are much more often found near each other in phage genomes, you might find some positive results that give you more confidence you have a helicase loader. Is this gene near other genes that look to be part of a replisome or near some type of helicase? If you find a nearby helicase then you have found additional evidence. Synteny is never strong evidence, but combining two pieces of weak evidence (synteny and partial HHPRED), can sometimes clearly provide sufficient evidence and give you confidence to "make the call".

Link to this post | posted 16 Mar, 2023 23:44
delesall	I'll add my two cents to what Chris wrote. I ended up NOT calling this a helicase loader in Rofo, mostly because there was no evidence for a helicase near by. In SPP1 and related phages, the pattern is helicase loader followed by helicae!

Link to this post | posted 29 Mar, 2023 12:30

EWisner

Thank you both for this information! For Culver (CQ1)the potential helicase loader (stop codon 62590 bp) is upstream for a gene we have annotated as DNA primase/helicase by ~3400 bps. The potential helicase loader gene is also directly downstream of a RepA-like replication initiator gene. With this information and the previously found evidence, I am inclined to call it the helicase loader but I wanted a second opinion on the matter. Does the helicase loader need to be directly followed by helicase, or just close? If just close by, then about how close?

Link to this post | posted 29 Mar, 2023 12:44
debbie	I would call this the helicase loader. I am not sure of the answer to your questions, but it is in the neighborhood. debbie