-
Notifications
You must be signed in to change notification settings - Fork 10
如何分析Gut microbiome与宿主Genome的互作
Ricky Woo edited this page Sep 27, 2017
·
8 revisions
如今,微生物组研究正如火如荼进行中,其中,微生物组与人类宿主基因组之间的互作的研究也是研究热点之一。那么,宿主基因组是如何影响微生物组的呢?
首先我们来看看这两种“组”分别有什么样的特点:
Human Microbiome | Human Genome |
---|---|
Many genomes | A single genome |
Recombination occurs independently of reproduction | Recombination occurs only during reproduction |
Functional capacity is dynamic | Basic functions are always present |
Changes in response to environment | Stable in response to the environment |
Highly mutable | Mutations are rare |
Generation time approx 30 min. | Generation time approx. 30 years |
如果单独对微生物组进行研究,一般的研究方案有:
- alpha diversity: 针对单个样本,样本中包含的物种(品系)的数量;
- beta diversity:不同样本微生物种群结构的差异
特定物种微生物的丰度,绝对丰度(abstract abundance)或者相对丰度(relative abundance)
针对微生物的功能进行分析
Taxonomic profiling | Functional profiling |
---|---|
High variation among individuals | Less variation among individuals |
Spatial and temporal colonization dynamics | Results of microbial interactions |
Key microbes associated with health | Key functions associated with health |
Selection, evolution, succession | Diagnosis, prognosis |
在菌株水平上(Strain-level)上分析序列的变异(sequence variation)以及基因表达水平(expression level)
常常不同的研究角度可能得出完全不同的结论,因而我们发现,微生物组的研究的可重复性相对较差。
- Jan Bonder et al. The effect of host genetics on the gut microbiome. Nature Genetics 48, 1407-1412 (2016)
Population
- 1514 individuals from 3 Dutch cohorts
-
Discovery cohort:
- LifeLines-DEEP cohort (984/1539 = 636 males + 903 females)
-
Replication cohort
- 500FG cohort (425/534 = 237 males + 296 females)
- MIBS-CO cohort (105/105 = 42 males and 63 females)
-
Discovery cohort:
Data
- Germline genotyping
- 8M SNPs targeted sequencing (explanatory variables)
- Gut mocrobiome shotgun sequencing
Covariates
- age
- gender
Outcome
- Microbial taxonomies (219 taxonomies)
- Functional units
- 636 MetaCyc pathways
- 661 Gene Ontoloty terms (2000 genes)
- Quality control (QC) of Reads
-
FastQC
,Trimmomatic
-
bowtie2
Filters the reads belonging to the human host genomes (NCBI38)
- Microbial compositional profiling
-
MetaPhlan 2.2
: 1M markers from > 7500 species
- Microbial compositional normalization
- The taxonomy data were normalized using an arcsin square root transformation
- Then corrected the normalized nonzero data for age, sex and read depth.
On the way to the garden of bioinformatics.
A bioinformatics wiki for the course BI462.