Author: Meixi Lin

Date: November 2018

This is a repo using a combination of anacapa and obitools to generate taxonomical classification for some NGS sequences.

1. move files: copy_rename_seq_10292018.sh
2. qc and demultiplex using anacapa: anacapa_qc_dada2_qsub.sh
3. ecotag classification:
   1. build & clean ref database:
      1. test_12SV5_ecopcr.sh
      2. clean_ecopcr.sh
   2. run classification: obitools_on_qcoutput_10302018.sh
4. anacapa classification:
   1. build crux database: build_crux_db_MAMM2.sh & a similar one
   2. create symlink for database used: create_symlink_crux_db.sh
   3. run classification: anacapa_classifier_qsub.sh

please refer to the config file

1. for obitools: https://git.metabarcoding.org/explore/projects
2. for anacapa: https://github.com/limey-bean/Anacapa

Date: 11/27/2018

Remember to set BLCAperMINlen="0.8"in the anacapa_vars.sh

In QC part

-m 30 instead of default -m 100 , here the minimum length after quality trimming is 30 instead of 100.

In run_bowtie2_blca.sh

-b percent similarity == 0.8 (this is default)

-p percent length match == 0.8 (this is default)

In

sequence similarity cutoff is 0.95

1. The CO1 was full length compared to default half-length
2. The length cutoff set for 12SV5 resulted in that some shorter sequences (e.g. 10 bps) were analyzed and the sequences were assigned taxonomy. these sequences need to be out.