Seqanswers Leaderboard Ad

Collapse

Announcement

Collapse
No announcement yet.
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Originally posted by ewingad View Post
    Would it also be valid use the -k 2 option and throw out reads for which two alignments are reported? This is slower than alignment against a masked genome but faster than -m 1.
    Absolutely, as long as you're using -k 2 in an unstratified reporting mode (the default in 0.10.0). Obviously, stratified -k 2 is not a good proxy for unstratified -m 1.

    I would be surprised if unstratified -k 2 performed all that differently from unstratified -m 1, since what's going on under the hood is essentially the same. Do you have an example where it is? If so, I should take a look.

    Ben

    Comment


    • Originally posted by Ben Langmead View Post
      Absolutely, as long as you're using -k 2 in an unstratified reporting mode (the default in 0.10.0). Obviously, stratified -k 2 is not a good proxy for unstratified -m 1.

      I would be surprised if unstratified -k 2 performed all that differently from unstratified -m 1, since what's going on under the hood is essentially the same. Do you have an example where it is? If so, I should take a look.

      Ben
      Actually now that I benchmark it, -m 1 is slightly faster than -k 2 using 0.10.0.

      -Adam

      Comment


      • Hi all,

        I'm using Bowtie to map some RNA-seq data, and I wanted to clarify my understanding of a couple points.

        The behaviour of -m 1 with default (0.10.0) parameters will only report results for which there is only one alignment anywhere within the 2-mismatch limit, right? So if there is an alignment with one mismatch and one with two, nothing will be reported. And if --strata is on, then the one-mismatch alignment will be reported (as long as there is only a single alignment with one mismatch). Is that all correct?

        Also, the rounding of quality values to between 10 and 30 means that there is no combination of two mismatches that give a total quality score of 70, so in effect the quality scores only affect the order of the results returned (which doesn't apply when -m 1 is on anyway). Have I got that right?

        Thanks a lot, and I apologize if any of this is explained in the manual or otherwise obvious.

        Kate

        Comment


        • Hi Kate,

          Originally posted by kcook View Post
          Hi all,
          The behaviour of -m 1 with default (0.10.0) parameters will only report results for which there is only one alignment anywhere within the 2-mismatch limit, right? So if there is an alignment with one mismatch and one with two, nothing will be reported. And if --strata is on, then the one-mismatch alignment will be reported (as long as there is only a single alignment with one mismatch). Is that all correct?
          Yes - that's all correct.

          Also, the rounding of quality values to between 10 and 30 means that there is no combination of two mismatches that give a total quality score of 70, so in effect the quality scores only affect the order of the results returned (which doesn't apply when -m 1 is on anyway). Have I got that right?
          The quality ceiling only applies in the -n ("Maq-like") alignment mode. So your statement is still correct for -v 2, but it's also the case that in -v 3 mode, alignments with combined mismatch qualities exceeding 70 are valid.

          I hope that helps,
          Ben

          Comment


          • Great! Thanks for the quick reply.

            Comment


            • Hello,

              Originally posted by kcook View Post
              Hi all,

              I'm using Bowtie to map some RNA-seq data, and I wanted to clarify my understanding of a couple points.

              The behaviour of -m 1 with default (0.10.0) parameters will only report results for which there is only one alignment anywhere within the 2-mismatch limit, right? So if there is an alignment with one mismatch and one with two, nothing will be reported. And if --strata is on, then the one-mismatch alignment will be reported (as long as there is only a single alignment with one mismatch). Is that all correct?

              Also, the rounding of quality values to between 10 and 30 means that there is no combination of two mismatches that give a total quality score of 70, so in effect the quality scores only affect the order of the results returned (which doesn't apply when -m 1 is on anyway). Have I got that right?

              Thanks a lot, and I apologize if any of this is explained in the manual or otherwise obvious.

              Kate
              kcook, you illuminated me! Now I anderstand the -m 1 better!

              Comment


              • bowtie 'hanging'

                Ben,

                It still seems to be doing this. When I look at the *map file, it obviously ends without properly finishing, as you can see from the last line.

                SOLEXA8_38_8_100_1783_191_0_2/2 - Castanea 105274 GATCCGTATCATCTTGACTTGGTTCTGATTTCTCTATTTTTTTAAGAATAC IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII 0
                SOLEXA8_38_8_100_1783_586_0_1/1 + Castanea 107330 CGTTACCTTAACCACAAGGAGGGGGATGCCGAAGGCAGGGCTAGTGACTGG IIIIII

                Originally posted by Ben Langmead View Post
                Hi Chuck,

                Please post the exact Bowtie version and arguments you're using. Also, please let me know if you see this problem when you use the latest version of Bowtie (0.10.0).

                Thanks,
                Ben
                I just downloaded the latest version last night (0.10.0.2).

                The job statement is as follows:

                ./bowtie -f Castmoll_cp -1 /media/upuna/LH0002/LH0002_302MJAAXX_1_1.fa,/media/upuna/LH0002/LH0002_3151AAAXX_2_1.fa,/media/upuna/LH0002/LH0002_3151AAAXX_1_1.fa,/media/upuna/LH0002/LH0002_3151AAAXX_3_1.fa -2 /media/upuna/LH0002/LH0002_302MJAAXX_1_2.fa,/media/upuna/LH0002/LH0002_3151AAAXX_2_2.fa,/media/upuna/LH0002/LH0002_3151AAAXX_1_2.fa,/media/upuna/LH0002/LH0002_3151AAAXX_3_2.fa /media/upuna/LH0002/LH2_all_castmoll.map

                Thanks,
                Chuck

                Comment


                • Hi Chuck,

                  Originally posted by chuck View Post
                  Ben,

                  It still seems to be doing this. When I look at the *map file, it obviously ends without properly finishing, as you can see from the last line.
                  Could you try the same command but using 'bowtie-debug' instead of 'bowtie'? If possible, pick a set of parameters where (a) 'bowtie' hangs and (b) the run doesn't take very long.

                  Thanks,
                  Ben

                  Comment


                  • Originally posted by Ben Langmead View Post

                    Could you try the same command but using 'bowtie-debug' instead of 'bowtie'? If possible, pick a set of parameters where (a) 'bowtie' hangs and (b) the run doesn't take very long.
                    Ben,

                    I don't seem to have a ./bowtie-debug command - I made this from the source code on a 64-bit machine.

                    Chuck

                    Comment


                    • Originally posted by chuck View Post
                      Ben,

                      I don't seem to have a ./bowtie-debug command - I made this from the source code on a 64-bit machine.

                      Chuck
                      Hi Chuck - sorry, just do 'make bowtie-debug' and that should create it.

                      Ben

                      Comment


                      • oops, my bad -

                        okay, it aborts very quickly using 'bowtie-debug'. Using 'bowtie', it almost finishes but seems to stop working as it approaches the end.

                        --results using 'debug' below

                        command>>> ./bowtie-debug -f Castmoll_cp -1 /media/upuna/TD0001/TD0001_3108EAAXX_7_1.fa -2 /media/upuna/TD0001/TD0001_3108EAAXX_7_2.fa /media/upuna/TD0001/TD1_all_castmoll_debug.map

                        RESULT>>>
                        Warning: Read (SOLEXA5_68_7_1_25_2044_0_1/1) is less than 3 characters long; skipping...
                        assert_gt: expected (0) > (0)
                        ebwt_search_backtrack.h:3265
                        bowtie-debug: ebwt_search_backtrack.h:3265: virtual void EbwtSeededRangeSourceDriver::setQueryImpl(PatternSourcePerThread*, Range*): Assertion `0' failed.
                        Aborted

                        Comment


                        • Hi everyone.

                          I would appreciate to have your comments on the following : when aligning the solexa reads with bowtie,
                          if a read aligns to multiple genomic regions, is the highest-scored location picked up in the final report
                          (i.e. when using --best option) ? And if a read aligns with the same score to multiple regions, would it
                          be possible to see the score of the alignment and the differences in the score among multiple regions ?
                          In this last scenario, a randomly picked location among the equally scored genomic locations is reported ?

                          thanks very much,

                          bogdan

                          Comment


                          • Hi Bogdan,

                            Originally posted by Bogdan Tanasa View Post
                            I would appreciate to have your comments on the following : when aligning the solexa reads with bowtie,
                            if a read aligns to multiple genomic regions, is the highest-scored location picked up in the final report
                            (i.e. when using --best option) ? And if a read aligns with the same score to multiple regions, would it
                            be possible to see the score of the alignment and the differences in the score among multiple regions ?
                            In this last scenario, a randomly picked location among the equally scored genomic locations is reported ?
                            I think your questions will be best answered by referring you to the reporting modes section of the manual, which includes some example invocations of Bowtie. Hope that helps,

                            Ben

                            Comment


                            • Hi Chuck,

                              If you have the time, there's one more thing it would help to try: remove the 'bowtie' binary and rebuild it using 'make EXTRA_FLAGS="-O1" bowtie', then re-run the problematic command using the new binary.

                              Thanks for your patience,
                              Ben

                              Comment


                              • Originally posted by Ben Langmead View Post
                                remove the 'bowtie' binary and rebuild it using 'make EXTRA_FLAGS="-O1" bowtie', then re-run the problematic command using the new binary.

                                Thanks for your patience,
                                Ben
                                doing this now.

                                I wanted to say that you are the patient one. I know how much work this software development is, particularly in the open environment!

                                All the best,
                                Chuck

                                Comment

                                Latest Articles

                                Collapse

                                • seqadmin
                                  Choosing Between NGS and qPCR
                                  by seqadmin



                                  Next-generation sequencing (NGS) and quantitative polymerase chain reaction (qPCR) are essential techniques for investigating the genome, transcriptome, and epigenome. In many cases, choosing the appropriate technique is straightforward, but in others, it can be more challenging to determine the most effective option. A simple distinction is that smaller, more focused projects are typically better suited for qPCR, while larger, more complex datasets benefit from NGS. However,...
                                  10-18-2024, 07:11 AM
                                • seqadmin
                                  Non-Coding RNA Research and Technologies
                                  by seqadmin




                                  Non-coding RNAs (ncRNAs) do not code for proteins but play important roles in numerous cellular processes including gene silencing, developmental pathways, and more. There are numerous types including microRNA (miRNA), long ncRNA (lncRNA), circular RNA (circRNA), and more. In this article, we discuss innovative ncRNA research and explore recent technological advancements that improve the study of ncRNAs.

                                  Nobel Prize for MicroRNA Discovery
                                  This week,...
                                  10-07-2024, 08:07 AM

                                ad_right_rmr

                                Collapse

                                News

                                Collapse

                                Topics Statistics Last Post
                                Started by seqadmin, Yesterday, 05:31 AM
                                0 responses
                                10 views
                                0 likes
                                Last Post seqadmin  
                                Started by seqadmin, 10-24-2024, 06:58 AM
                                0 responses
                                20 views
                                0 likes
                                Last Post seqadmin  
                                Started by seqadmin, 10-23-2024, 08:43 AM
                                0 responses
                                48 views
                                0 likes
                                Last Post seqadmin  
                                Started by seqadmin, 10-17-2024, 07:29 AM
                                0 responses
                                58 views
                                0 likes
                                Last Post seqadmin  
                                Working...
                                X