Based on the feedbacks from the Centrifuge development team, I removed Bracken post-processing of Centrifuge result, but just used the Kraken-style report directly produced by Centrifuge as the final result. This report files contains read counts and relative abundances per taxon.
Therefore I also added the combine_kraken function. This function handles the Kraken-style reports. In oecophylla, it actually processed Centrifuge output instead of Kraken output.
Meanwhile, the old combine_centrifuge function is retained, because it directly handles the native Centrifuge output file (not Kraken-style report). This native output file contains read counts and relative abundances per genome, which is not what we want.
qiyunzhu
commented
7 years ago
Based on the feedbacks from the Centrifuge development team, I removed Bracken post-processing of Centrifuge result, but just used the Kraken-style report directly produced by Centrifuge as the final result. This report files contains read counts and relative abundances per taxon.
Therefore I also added the
combine_krakenfunction. This function handles the Kraken-style reports. In oecophylla, it actually processed Centrifuge output instead of Kraken output.Meanwhile, the old
combine_centrifugefunction is retained, because it directly handles the native Centrifuge output file (not Kraken-style report). This native output file contains read counts and relative abundances per genome, which is not what we want.Also tested the recently release Kraken 1.0.
Addressed issues #121 and #122