Kraken | Homology-Based Taxonomic

[…] million reads per library., raw sequencing reads were visualized with fastqc (http://www.bioinformatics.babraham.ac.uk/projects/fastqc/) to determine the necessary quality filtering steps. we used kraken [22] to eliminate any potential bacterial or viral contamination, and sortme [23] to remove reads that were rrna rather than mrna to avoid misannotation of rrnas as proteins [24]. low quality […]

[…] 0.36 (31) to trim off low-quality reads and bowtie 2.2.9 (32) against human reference genome assembly grch38 to remove potential human contamination. taxonomic profiling was conducted with kraken 0.10.5 (33) against a database that contains all of the complete genomes in the ncbi refseq database under the categories bacteria, archaea, viral, protozoa, fungi, and human as of november […]

[…] trimmomatic v_0.36 was used to remove low quality sequences. human sequences were removed after alignment with reference genome (hg38 database [41]) using bowtie2 v_2.2.9, with default settings. kraken v_0.10.5-beta was then used to for taxonomic classification of unmapped microbial reads [42]. by the default setting of kraken, only complete genome sequences were included as the reference […]

[…] (hg38 database [41]) using bowtie2 v_2.2.9, with default settings. kraken v_0.10.5-beta was then used to for taxonomic classification of unmapped microbial reads [42]. by the default setting of kraken, only complete genome sequences were included as the reference genome database to reduce the errors from contaminants [42]. since our project focused on bacteria community, we only included […]

[…] (materials and methods) (42). as expected from a gut metagenome, the major clades in the sample were firmicutes (70%) and bacteroidetes (15%) (fig. 1b) (43). an analysis of dna sequences using the kraken program (44) (materials and methods) confirmed these results with, also as expected, ~80% of the sequences not attributed to a specific genome (43). coding sequences were then annotated […]