使用blast+megan 来分析物种

anneng

1.使用blast来比对（这一步很慢需要用我们的大数据方案）
nohup blastn -query barcode05.fasta -db /ceph_disk1/gene_data/MetaDatabase/NCBI_blast_db_FASTA/nt/ntdata/nt -num_threads 20 -out barcode05.m8 -outfmt 6 -evalue 0.001 &
2.准备工作：制作序列id和taxid的映射文件把ncbi的映射文件转成megan自己的格式
在MEGAN.vmoptions 文件中配置java的最大内存：
-Xmx80000M 否则会报内存不足

./make-acc2ncbi -i nucl_gb.accession2taxid.gz 
Version   MEGAN Ultimate Edition (version 6.21.2, built 14 Mar 2021)
Author(s) Daniel H. Huson
Copyright (C) 2020 Daniel H. Huson. This program comes with ABSOLUTELY NO WARRANTY.
Computing map:
Processing file: nucl_gb.accession2taxid.gz
10% 20% 30% 40% 100% (193.8s)
Building table:
(Bits: 26, buckets: 67,108,864, bucket size: 4)
Sorting map...
10% 20% 30% 40% 50% 60% 70% 80% 90% 100% (210.1s)
Writing table...
10% 20% 100% (38.7s)
(Bucket avg size: 3.9, max size: 19, used: 98%)
(Index size:  536,870,916, data size: 3,603,383,706)
Merging files
10% 100% (42.0s)
Opening file: acc2tax.abin
Size:  259,629,219
Total in:  259,629,220 
Total out: 259,629,219 
Total time:  497s
Peak memory: 60.2 of 78.1G

生成的m8 文件可以直接到入megan分析不过文件都比较大在PC机上分析很慢我们还是建议用户把数据上传到我们服务器进行分析

3.用megan的blast2lca 来注释物种

/opt/megan-ce/tools/blast2lca -i barcode05.m8.2 -f BlastTab -m BlastN -o barcode05.lca.norank -a2t /opt/megan-ue/tools/ncbi/acc2tax.abin -sr false

生成的例子如下：这个格式我们要用图表展示到UI上
f3d2e450-007a-4c9e-b43b-602e6302df32; ; Eukaryota; 100; Metazoa; 100; Chordata; 100; Mammalia; 100; Primates; 100; Hominidae; 50; Homo; 50; Homo sapiens; 50;

常见错误：
报错格式不对
/opt/megan-ce/tools/blast2lca -i barcode05.m8 -f BlastTab -m BlastN -o barcode05.lca -a2t ~/.basta/taxonomy/nucl_gb.accession2taxid
Warning: Might not be a BLAST file in TAB format: barcode05.m8
Error parsing file near line: 12028904: String index out of range: 53
Error parsing file near line: 12028905: String index out of range: 53
Error parsing file near line: 17258173: String index out of range: 53

anneng

blast可以直接给出物种名称但是要注意加一个全局变量
ad1bd439-6b2b-4178-91d5-8c6f36fd8d6f,N/A,58.4

物种显示NA 需要在bashrc中添加一个全局变量
export BLASTDB=/ceph_disk1/gene_data/MetaDatabase/NCBI_blast_db_FASTA/nt/ntdata/

nohup blastn -query barcode05.fasta -db /ceph_disk1/gene_data/MetaDatabase/NCBI_blast_db_FASTA/nt/ntdata/nt -num_threads 20 -out barcode05.m8 -outfmt "6 qseqid,sscinames,bitscore" -evalue 0.001 &

anneng

下面这个工具也实现了megan lca相关的算法　可以重点关注
https://github.com/fungs/taxator-tk/
https://github.com/emepyc/Blast2lca (已过时)
https://github.com/etheleon/pymegan https://etheleon.github.io/articles/pythonMEGAN/

https://www.biostars.org/p/362985/ blast2lca的解析
https://www.jianshu.com/p/3d3253c59545 2022-03-18 python pandas 处理megan中blast2lca的结果统计

anneng

The program make-acc2ncbi can be used to create a new accession to taxonomy mapping file from files downloaded from ftp://ftp.ncbi.nih.gov/pub/taxonomy/accession2taxid.
这个工具只在Megan UE版本中有

anneng

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5978398/
BLAST-based validation of metagenomic sequence assignments
使用blast进行物种注释
在法医场景下精度会非常重要这篇文章对低精度软件的分析结果使用blast进行二次验证

anneng

NCBI Blast output header -outfmt 6 or -m8(新版本没有这个选项了) header in tabular form

query_id subject_id pct_identity aln_length n_of_mismatches gap_openings q_start q_end s_start s_end e_value bit_score

anneng

http://megan.informatik.uni-tuebingen.de/t/problems-importing-blast-output-into-megan/1178

ice-melt

准备工作制作序列id和taxid的映射文件

下载地址：
https://ftp.ncbi.nih.gov/pub/taxonomy/accession2taxid/

prot.accession2taxid.gz
nucl_gb.accession2taxid.gz

使用blast+megan 来分析物种

准备工作 制作序列id和taxid的映射文件

准备工作制作序列id和taxid的映射文件