de novo RNA loop

Member Site › Forums › Rosetta 3 › Rosetta 3 – General › de novo RNA loop

Viewing 3 reply threads
  • Author
    Posts
    • February 15, 2021 at 2:34 am #3696
      Anonymous

        Hello, 

        I’m using the rosetta rna de novo tool to generate a rna hairpin, I used the RNA_Denovo_with_base_pair_steps demo as a guide to do it, however I get the error: Atom C1′ 1 not found. I’ m newbie using rosetta so I would really appreciate any help or guide about the problem. Thank you!

      • February 16, 2021 at 5:19 am #15747
        Anonymous

          Can you share the inputs you provided to the executable? It looks to me like it’s quite confused by your requested sequence.

          • February 16, 2021 at 2:39 pm #15748
            Anonymous

              Sure.. I used the command /hpcfs/apps/rosetta/3.7/main/source/bin/rna_denovo.linuxgccrelease @flags

              Flags contains these parameters:

              -fasta RiboON.fasta

              -nstruct 1

              -cycles 5000 

              -filter_lores_base_pairs_early 

              -bps_moves

              -output_filters 

              -output_lores_silent_file

              -score:rna_torsion_potential RNA11_based_new

              -obligate_pair  1 40  2 39  3 38  4 37  5 36  6 35  7 34  8 33  9 32  10 31  11 30  12 26  13 25  14 24  15 23  16 22  47 69  48 68  49 67  50 66  51 65  52 64  53 62  54 61  55 60 

              -minimize_rna

              -out:file:silent RiboON.out

              and RiboON.fasta have this information:

              >RiboON strand 1

              AAGUAAAGGCAAUAUCGUCUUGAUAUACGUAAGUGCACUCCAUUUACAUACUCGGUAAACUGGAGUGCACUUACUUUUUUUUAGGAGGUUAAUGAUC

              Thanks in advance!!

          • February 24, 2021 at 5:31 am #15757
            Anonymous

              OK, the issue is kind of multifold.

              For one thing, your fasta should use lowercase letters for RNA, so it can tell the difference from proteins. It should also have the numbering in the title line — something like “>RiboON_strand_1 1-98” or whatever the sequence length is instead of 98. Right now I think it’s failing catastrophically to parse your silent file.

              In general, though, I think you would do well to follow along some of the examples here: https://www.rosettacommons.org/docs/latest/FARFAR2

              Instead of a long -obligate_pair flag you could probably just specify dot-bracket secondary structure like

              ((((((((((((((((…..))))))))))))))))……(((((((((….)))))))))…………………………..

              (if I interpret the length correctly)

              Separately, you’re using a relatively old version of Rosetta, and a lot has changed since 2016. I really suggest upgrading!

               

              Finally, please feel free to submit this job to the webserver at https://rosie.rosettacommons.org/farfar2

            • March 18, 2021 at 2:13 am #15798
              Anonymous

                Hello

                Thank you very much for your reply. I made the changes that you told me but now I have a complementarity error in some bases. To solve this I had used the -obligate_pair flag, but I don’t know if this is the best way. Is there another way to solve this error?

                Thanks in advance for your help

                • March 30, 2021 at 3:19 am #15832
                  Anonymous

                    Good question. Are you sure your secondary structure is correct? A complementarity error means you’re implying there’s (for example) a GA base pair.

                    If you like, you can specify your -secstruct (or -secstruct_file) — which implies all WC, canonical pairing — as well as a -secstruct_general (or -secstruct_general_file), where all the complementarity requirements are removed.

              • Author
                Posts
              Viewing 3 reply threads
              • You must be logged in to reply to this topic.