SEA-PHAGES | All posts created by Pollenz

Link to this post \| posted 25 May, 2023 12:25
Pollenz	We have use a Bioinformatic approach to identify several phage clusters that have a distal holin-like gene cassette that are not associated with the lysins. This type of organization can be observed in the B1, B2, B3, B5, B9, DR and DO phages. Most of the cassettes have four different TM genes with a 4-4-4-1 configuration and in many cases all these genes have a 4bp overlap to each other. We have bioinformatically looked at all of the pham members for all of these distal holin-like genes and CAN FIND THEM DIRECTLY DOWNSTREAM of lysins in DD, EP, DV, DG1, DK CZ2, DB, CS4, CV, CY and CZ7. This clearly validates that they are likely functioning in lysis. I have enclosed an colorful analysis based on the 4 genes in the DR Phage CaiB. Each color represents a protein in the same pham. In the case of your B1 phages, note that phages in the DD cluster have both 4TM genes directly downstream of a split lysin followed by a 1TM gene. Based on our wet lab recent work, it is likely that ALL of the TM genes are holins (including the 1TM that may be the most important one. DATA IN PREP but presented at 2023 Phages Conference). Note that the majority of these multi-TM gene cassettes have a terminal 1TM gene. See our PLOSONE paper as well for a comprehensive analysis of these types of TM cassettes. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0276603 RS Pollenz Edited 25 May, 2023 15:43 1011Kb

Link to this post | posted 25 May, 2023 12:25

We have use a Bioinformatic approach to identify several phage clusters that have a distal holin-like gene cassette that are not associated with the lysins. This type of organization can be observed in the B1, B2, B3, B5, B9, DR and DO phages. Most of the cassettes have four different TM genes with a 4-4-4-1 configuration and in many cases all these genes have a 4bp overlap to each other.

We have bioinformatically looked at all of the pham members for all of these distal holin-like genes and CAN FIND THEM DIRECTLY DOWNSTREAM of lysins in DD, EP, DV, DG1, DK CZ2, DB, CS4, CV, CY and CZ7. This clearly validates that they are likely functioning in lysis. I have enclosed an colorful analysis based on the 4 genes in the DR Phage CaiB. Each color represents a protein in the same pham.

In the case of your B1 phages, note that phages in the DD cluster have both 4TM genes directly downstream of a split lysin followed by a 1TM gene. Based on our wet lab recent work, it is likely that ALL of the TM genes are holins (including the 1TM that may be the most important one. DATA IN PREP but presented at 2023 Phages Conference). Note that the majority of these multi-TM gene cassettes have a terminal 1TM gene. See our PLOSONE paper as well for a comprehensive analysis of these types of TM cassettes. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0276603

RS Pollenz

Edited 25 May, 2023 15:43

Posted in: Cluster B Annotation Tips → Holin

Link to this post \| posted 11 Feb, 2023 22:52
Pollenz	We are annotating phage Azira. It has high identity to survivors. In Azira, it is clear that gene #66 loses coding potential 1/2 way through the ORF, most likely due to losing a stop site. Gene #67 has coding potential across the entire ORF and this gene in most PHAM members is annotated with the longer ORF. Calling the LORF results in a 121bp overlap to gene #66 and this is what has been annotated for Survivors. All other phages that have this gene, do NOT have long overlaps to the previous gene as they all have smaller genes that precede it. Its obvious that Azira gp67 should be annotated to include the full ORF, especially since gp66 has no CP right at the point that gp67 starts, just as Survivors, but this highly violates the gene overlap rules. Anyone have any thoughts on this. I enclosed the Genmark map and you can see the situation, Azira gp67 is the gene starting at 44,000. As an added intrigue to this situation, there is a start site at 44161 for gp67 with a -4bp overlap to the end of gp66. Calling this results in a 93bp ORF and 31aa protein. Also violations of the general rule that ORFS are usually >120bp…. RS Pollenz 75Kb

Link to this post | posted 11 Feb, 2023 22:52

Pollenz

We are annotating phage Azira. It has high identity to survivors. In Azira, it is clear that gene #66 loses coding potential 1/2 way through the ORF, most likely due to losing a stop site. Gene #67 has coding potential across the entire ORF and this gene in most PHAM members is annotated with the longer ORF. Calling the LORF results in a 121bp overlap to gene #66 and this is what has been annotated for Survivors. All other phages that have this gene, do NOT have long overlaps to the previous gene as they all have smaller genes that precede it.

Its obvious that Azira gp67 should be annotated to include the full ORF, especially since gp66 has no CP right at the point that gp67 starts, just as Survivors, but this highly violates the gene overlap rules.

Anyone have any thoughts on this. I enclosed the Genmark map and you can see the situation, Azira gp67 is the gene starting at 44,000.

As an added intrigue to this situation, there is a start site at 44161 for gp67 with a -4bp overlap to the end of gp66. Calling this results in a 93bp ORF and 31aa protein. Also violations of the general rule that ORFS are usually >120bp….

RS Pollenz

Posted in: Cluster CT Annotations Tips → BIG GENE OVERLAP

Link to this post \| posted 21 Dec, 2022 21:01
Pollenz	Great work on this question. The ribbon-helix-helix is a DNA binding motif found in many proteins. It is typified by a B-sheet of 6-10 amino acids followed by a short spacer and two alpha helices. It can be found in many locations within a primary sequence of a protein, but must contain all THREE domains. In HHpred the alpha helices are noted by Hh and beta sheets by Ee in the ss_pred line of the alignments. Capital letters indicate high confidence in the secondary structure (ss) prediction. See PDB 6SBW_A for a good schematic of this type of ribbon HH domain (enclosed schematic). Note that Usavi70 has alignments to the subjects with Q1 to all hits. Amino acid 1 is the start of a helix and there is also a second helix in Usavi70, but the Usavi70 has NO beta sheet that precedes the hits compared to the queries. See that the deeper analysis of the PDB queries show that they are all aligning well past amino acid 1 and all have a 7-9 amino acid beta sheet that precedes the helix 1 that is aligned to Usavi70. Thus, the ribbon helix helix IS NOT supported for Usavi70 and should not be called. NCBI conserved domains also does not support either an HTH or DNA binding call as well. So best to leave a hypothetical NKF since all components of the domain are not present. Hope this is helpful RS Pollenz 485Kb

Link to this post | posted 21 Dec, 2022 21:01

Pollenz

Great work on this question.

The ribbon-helix-helix is a DNA binding motif found in many proteins. It is typified by a B-sheet of 6-10 amino acids followed by a short spacer and two alpha helices. It can be found in many locations within a primary sequence of a protein, but must contain all THREE domains. In HHpred the alpha helices are noted by Hh and beta sheets by Ee in the ss_pred line of the alignments. Capital letters indicate high confidence in the secondary structure (ss) prediction.

See PDB 6SBW_A for a good schematic of this type of ribbon HH domain (enclosed schematic).

Note that Usavi70 has alignments to the subjects with Q1 to all hits. Amino acid 1 is the start of a helix and there is also a second helix in Usavi70, but the Usavi70 has NO beta sheet that precedes the hits compared to the queries. See that the deeper analysis of the PDB queries show that they are all aligning well past amino acid 1 and all have a 7-9 amino acid beta sheet that precedes the helix 1 that is aligned to Usavi70.

Thus, the ribbon helix helix IS NOT supported for Usavi70 and should not be called. NCBI conserved domains also does not support either an HTH or DNA binding call as well. So best to leave a hypothetical NKF since all components of the domain are not present.

Hope this is helpful

RS Pollenz

Posted in: Functional Annotation → Usavi_draft_70: potentially ribbon-helix-helix?

Link to this post \| posted 24 May, 2022 18:44
Pollenz	The DR Cluster phages appear to have a 3-5 gene transposable operon with -4bp overlaps at the 5' end of the genome prior to the terminase gene. There are several variations of the operon across all the phages with some shared genes and ouliers like Evaa and Sour that are clearly divergent from the majority of cluster members. Most genes within these "operons" have HHpred hits, but specific functional calls to some of the genes can not be made as the hits do not match to those on the Official genes list. These operons appears to encode genes involved in nucleotide modification and may be involved in phage defense against restriction as noted in Nucleic Acids Research, 2013, Vol. 41, No. 16 7635–7655. Note that the 5 gene operons contains a possible adenylate kinase gene that is very tricky and requires deep review of the HHpred hits as finding the phosphate-binding loop (P-loop) that is a conserved sequence motif found in mononucleotide-binding proteins/kinases and DOES NOT specify a specific kinase (Ex. compare the hits of CaiB_4 to LittleMunchkin_4). Adenylate kinases do have crystal structures in HHpred (2C95_B, 2BWJ_B, 3UMF_A) and cd hits in NCBI so there should be evidence beyond the P-loop to make the call. See also the notes on Gene 1 from Cluster B. RS Pollenz Edited 24 May, 2022 20:09

Link to this post | posted 24 May, 2022 18:44

Pollenz

The DR Cluster phages appear to have a 3-5 gene transposable operon with -4bp overlaps at the 5' end of the genome prior to the terminase gene. There are several variations of the operon across all the phages with some shared genes and ouliers like Evaa and Sour that are clearly divergent from the majority of cluster members. Most genes within these "operons" have HHpred hits, but specific functional calls to some of the genes can not be made as the hits do not match to those on the Official genes list. These operons appears to encode genes involved in nucleotide modification and may be involved in phage defense against restriction as noted in Nucleic Acids Research, 2013, Vol. 41, No. 16 7635–7655. Note that the 5 gene operons contains a possible adenylate kinase gene that is very tricky and requires deep review of the HHpred hits as finding the phosphate-binding loop (P-loop) that is a conserved sequence motif found in mononucleotide-binding proteins/kinases and DOES NOT specify a specific kinase (Ex. compare the hits of CaiB_4 to LittleMunchkin_4). Adenylate kinases do have crystal structures in HHpred (2C95_B, 2BWJ_B, 3UMF_A) and cd hits in NCBI so there should be evidence beyond the P-loop to make the call. See also the notes on Gene 1 from Cluster B.

RS Pollenz

Edited 24 May, 2022 20:09

Posted in: Cluster DR Annotation Tips → Metabolic Operon

Link to this post \| posted 20 May, 2022 19:20
Pollenz	Many of the G1 phages are highly conserved at the nucleotide level in the region between the last reverse gene (immunity repressor) and first forward gene (cro). The operator site can be used to guide the annotation and is 5'-CGACATATGTCG-3'. Note that wet lab data supports the identification of the putative -10 and -35 binding sites and this orients the location of the CORRECT start codon (see the paper referenced above). In some cases, the DNA master RBS scores for the correct start site ARE NOT VERY GOOD and the start may not have been chosen by Genemark or Glimmer. However the wet lab data from BPs should be used to annotate the proper starts. Note that this results in smaller protein products, but DOES NOT reduce HHpred hits to the relevant PDBs or truncate N-terminal amino acid sequences. Remember that the intergenic region when moving from a reverse to forward gene MUST contain a sufficient gap (usually of at least 50bp) to accommodate these regulatory regions for transcription of these genes. RS Pollenz

Link to this post | posted 20 May, 2022 19:20

Pollenz

Many of the G1 phages are highly conserved at the nucleotide level in the region between the last reverse gene (immunity repressor) and first forward gene (cro). The operator site can be used to guide the annotation and is 5'-CGACATATGTCG-3'. Note that wet lab data supports the identification of the putative -10 and -35 binding sites and this orients the location of the CORRECT start codon (see the paper referenced above). In some cases, the DNA master RBS scores for the correct start site ARE NOT VERY GOOD and the start may not have been chosen by Genemark or Glimmer. However the wet lab data from BPs should be used to annotate the proper starts. Note that this results in smaller protein products, but DOES NOT reduce HHpred hits to the relevant PDBs or truncate N-terminal amino acid sequences. Remember that the intergenic region when moving from a reverse to forward gene MUST contain a sufficient gap (usually of at least 50bp) to accommodate these regulatory regions for transcription of these genes.

RS Pollenz

Posted in: Cluster G Annotation Tips → integration cassette

Link to this post \| posted 20 May, 2022 17:06
Pollenz	I have done some analysis of RecE while annotating G1 phage ShaboiShabazz gp42. First, the PDB hit to 3H4R_A are to an E. coli exodeoxyribonuclease VIII, this is the putative RecE gene and has no connection to Cas4. Second, the publication regarding this RecE gene has identified several “conserved” domains and the genes within the PHAM that includes ShaboiShabazz gp42 contain 4 of these domains with near 100% conservation of not only the key amino acids, but also the secondary structures (B sheets and alpha helices). Third, RecE (exodeoxyribonuclease VIII) function in concert with the RecT. RecE binds to free double-stranded DNA(dsDNA) ends and processively digests the 50-endedstrand to form 50-mononucleotides and a 30-over-hang that is a substrate for single strand annealing promoted by RecT (2009: Structure 17, 690–702). ShaboiShabazz gp43 hits to PF03837.17: RecT family. Its key to note that RecT functions in BOTH RecA-dependent and RecA-independent DNA recombination pathways, so having a hit to this Pfam is not specific to Cas4 and I’m not sure how these connections were even made. **SEE THE POSTED ANALYSIS OF gp42 showing the clear hits to RecE. For Pfam 12684.10: This entry represent a PD-(D/E)XK endonuclease-like domain superfamily. PD-(D/E)XK nucleases constitute a large and highly diverse superfamily of enzymes that display little sequence similarity. However, they share a common core fold and a few critical active site residues and ARE NOT specific to Cas4 either. Makes sense for RecE to hit to this Pfam based on overall function. Thus, hits to 3H4R_A SHOULD CALL RecE and having a downstream RecT also confirm that calling both genes is valid. Thus: ShaboiShabazz gp42 is RecE-like exonuclease and gp43 is RecT-like ssDNA binding protein. RS Pollenz Edited 20 May, 2022 17:07 1.77Mb

Link to this post | posted 20 May, 2022 17:06

Pollenz

I have done some analysis of RecE while annotating G1 phage ShaboiShabazz gp42. First, the PDB hit to 3H4R_A are to an E. coli exodeoxyribonuclease VIII, this is the putative RecE gene and has no connection to Cas4. Second, the publication regarding this RecE gene has identified several “conserved” domains and the genes within the PHAM that includes ShaboiShabazz gp42 contain 4 of these domains with near 100% conservation of not only the key amino acids, but also the secondary structures (B sheets and alpha helices). Third, RecE (exodeoxyribonuclease VIII) function in concert with the RecT. RecE binds to free double-stranded DNA(dsDNA) ends and processively digests the 50-endedstrand to form 50-mononucleotides and a 30-over-hang that is a substrate for single strand annealing promoted by RecT (2009: Structure 17, 690–702). ShaboiShabazz gp43 hits to PF03837.17: RecT family. Its key to note that RecT functions in BOTH RecA-dependent and RecA-independent DNA recombination pathways, so having a hit to this Pfam is not specific to Cas4 and I’m not sure how these connections were even made. **SEE THE POSTED ANALYSIS OF gp42 showing the clear hits to RecE.

For Pfam 12684.10: This entry represent a PD-(D/E)XK endonuclease-like domain superfamily. PD-(D/E)XK nucleases constitute a large and highly diverse superfamily of enzymes that display little sequence similarity. However, they share a common core fold and a few critical active site residues and ARE NOT specific to Cas4 either. Makes sense for RecE to hit to this Pfam based on overall function.

Thus, hits to 3H4R_A SHOULD CALL RecE and having a downstream RecT also confirm that calling both genes is valid. Thus: ShaboiShabazz gp42 is RecE-like exonuclease and gp43 is RecT-like ssDNA binding protein.

RS Pollenz

Edited 20 May, 2022 17:07

Posted in: Functional Annotation → Refining the call for Cas4 family exonuclease vs. RecE-like exonuclease

Link to this post \| posted 09 May, 2022 16:00
Pollenz	In the DJ phages (and also in many of the Gordonia clusters) there is evidence for numerous intergenic 50-75bp repeated sequence are are located directly upstearm of the ATG start sites for 7-10 genes in the middle of the genome, (in the DJ the area is just upstream of the tape measure gene). Note that these areas are easily identified when looking at the Phamerator maps as solid lines that diverge to numerous areas and hit the genome UPSTREAM of numerous genes. The nature of the sequence can be found using the SCAN function in DNA master and also using programs like CLUSTAL OMEGA to identify the alignments. They key is that although gaps of 50-75bp between genes are atypical, in some cases there are start sites that can be found within these regions that create longer ORFS. These longer starts should not be called and its important to assess both the Glimmer and Genmark calls as usually they are the correct start for these types of genes. In some cases the longer starts have been annotated and this creates confusion. RS Pollenz 1.31Mb

Link to this post | posted 09 May, 2022 16:00

Pollenz

In the DJ phages (and also in many of the Gordonia clusters) there is evidence for numerous intergenic 50-75bp repeated sequence are are located directly upstearm of the ATG start sites for 7-10 genes in the middle of the genome, (in the DJ the area is just upstream of the tape measure gene). Note that these areas are easily identified when looking at the Phamerator maps as solid lines that diverge to numerous areas and hit the genome UPSTREAM of numerous genes. The nature of the sequence can be found using the SCAN function in DNA master and also using programs like CLUSTAL OMEGA to identify the alignments. They key is that although gaps of 50-75bp between genes are atypical, in some cases there are start sites that can be found within these regions that create longer ORFS. These longer starts should not be called and its important to assess both the Glimmer and Genmark calls as usually they are the correct start for these types of genes. In some cases the longer starts have been annotated and this creates confusion.

RS Pollenz

Posted in: Cluster DJ Annotation Tips → Intergenic Repeated Sequences

Link to this post \| posted 08 Feb, 2022 02:05
Pollenz	Its been ~2 years since this thread and there has been several different scenarios regarding the tail assembly chaperone (TAC) in DR phages. There are currently 8 annotated phages with 5 different TAC annotations (see enclosed file). All of the DR phages have high nucleotide similarity and the same set of genes upstream of the tape measure. All of the genes are from huge PHAMS with 350-550 members across multiple clusters. The 3rd gene upstream of the tape measure has a majority call of hypothetical, only 3 DR phages call this TAC out of 406 members. The 2nd gene upstream has a MAJORITY call of TAC and the gene directly upstream of the tape measure has 55 of 384 calls to TAC. The issue is that in the DR phages, there are FIVE different annotation scenarios for the 8 annotated genomes in this region: 1) No TAC calls to any of the genes 2) TAC call to only the 2nd and 3rd upstream genes (Mariokart) 3) TAC calls with programmed frameshift between 2nd and 3rd upstream genes (Sour and NHagos) 4) TAC call to only the 2nd upstream gene (AnClar) 5) TAC with programmed frameshift between 1st and 2nd upstream genes (CloverMinnie) We are annotating CaiB. We cant have a programmed frameshift between the 2nd and 3rd upstream gens due to an in frame stop codon to the 2nd upstream gene that limits the sequence area that can be used for the slippage to ~25bp (and there is no slippery sequence within this region) We CAN find a non-canonical CGGGGCCG sequence in the 2nd upstream gene that would allow slippage to the 1st upstream gene (just as annotated in CloverMinnie with the SAME slippery sequence. Thoughts, ideas on this DR cluster? There should be consistency based on genes and sequences. Based on lack of wet lab data regarding the slippery sequences in Gordonia, seems that a non-canonical slippery call is preamature. Based on synteny and HHpred, seems that TAC should be called for 2nd upstream gene and possibly 1st upstream, but NOT for 3rd upstream. RS Pollenz 430Kb

Link to this post | posted 08 Feb, 2022 02:05

Pollenz

Its been ~2 years since this thread and there has been several different scenarios regarding the tail assembly chaperone (TAC) in DR phages. There are currently 8 annotated phages with 5 different TAC annotations (see enclosed file). All of the DR phages have high nucleotide similarity and the same set of genes upstream of the tape measure. All of the genes are from huge PHAMS with 350-550 members across multiple clusters. The 3rd gene upstream of the tape measure has a majority call of hypothetical, only 3 DR phages call this TAC out of 406 members. The 2nd gene upstream has a MAJORITY call of TAC and the gene directly upstream of the tape measure has 55 of 384 calls to TAC.

The issue is that in the DR phages, there are FIVE different annotation scenarios for the 8 annotated genomes in this region:

1) No TAC calls to any of the genes
2) TAC call to only the 2nd and 3rd upstream genes (Mariokart)
3) TAC calls with programmed frameshift between 2nd and 3rd upstream genes (Sour and NHagos)
4) TAC call to only the 2nd upstream gene (AnClar)
5) TAC with programmed frameshift between 1st and 2nd upstream genes (CloverMinnie)

We are annotating CaiB. We cant have a programmed frameshift between the 2nd and 3rd upstream gens due to an in frame stop codon to the 2nd upstream gene that limits the sequence area that can be used for the slippage to ~25bp (and there is no slippery sequence within this region)

We CAN find a non-canonical CGGGGCCG sequence in the 2nd upstream gene that would allow slippage to the 1st upstream gene (just as annotated in CloverMinnie with the SAME slippery sequence.

Thoughts, ideas on this DR cluster? There should be consistency based on genes and sequences. Based on lack of wet lab data regarding the slippery sequences in Gordonia, seems that a non-canonical slippery call is preamature. Based on synteny and HHpred, seems that TAC should be called for 2nd upstream gene and possibly 1st upstream, but NOT for 3rd upstream.

RS Pollenz

Posted in: Cluster DR Annotation Tips → Tail Assembly Chaperone

Link to this post \| posted 03 Dec, 2021 14:22
Pollenz	Using GC/MS of several EK2 phages, one of the most expressed proteins that is identified is the gene product for the gene directly downstream of the called portal protein. Although we do not have direct evidence that this is the major capsid, the location, similarity across all EK phages and its level of expression in a phage protein lysate suggest that this is a good candidate to be the MCP. We are also using cro-EM to begin to assess which protein is the MCP in these podoviriae phages. RS Pollenz

Link to this post | posted 03 Dec, 2021 14:22

Pollenz

Using GC/MS of several EK2 phages, one of the most expressed proteins that is identified is the gene product for the gene directly downstream of the called portal protein. Although we do not have direct evidence that this is the major capsid, the location, similarity across all EK phages and its level of expression in a phage protein lysate suggest that this is a good candidate to be the MCP. We are also using cro-EM to begin to assess which protein is the MCP in these podoviriae phages.

RS Pollenz

Posted in: Cluster EK Annotation Tips → major capsid protein

Link to this post \| posted 03 Jun, 2021 14:19
Pollenz	We have found in Phage Leperchaun (cluster F1) a split methyltransferase with very good HHpred evidence (gp68/gp69). In other cluster F1 there is one large methylase gene in this region, but in Leperchuan the first smaller ORF has a stop codon and the switch to frame 2 picks up the sequence with an 80bp overlap and a large second gene. In several of the HHpred hits, there are alignment to the same PDB that show continuous coverage between the 2 Leperchuan genes. There is no consensus slippery sequence but it will be interesting to see if this is the case. See attachment. RS Pollenz 1.59Mb

Link to this post | posted 03 Jun, 2021 14:19

Pollenz

We have found in Phage Leperchaun (cluster F1) a split methyltransferase with very good HHpred evidence (gp68/gp69). In other cluster F1 there is one large methylase gene in this region, but in Leperchuan the first smaller ORF has a stop codon and the switch to frame 2 picks up the sequence with an 80bp overlap and a large second gene. In several of the HHpred hits, there are alignment to the same PDB that show continuous coverage between the 2 Leperchuan genes. There is no consensus slippery sequence but it will be interesting to see if this is the case. See attachment.

RS Pollenz

Posted in: Annotation → Mid-gene deletion causing frameshift and orphams

Recent Activity

All posts created by Pollenz