Download 2014 Spring Seminar Speaker Series Xuegong Zhang, PhD

Survey
yes no Was this document useful for you?
   Thank you for your participation!

* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project

Document related concepts

Biology and consumer behaviour wikipedia , lookup

Genetic studies on Bulgarians wikipedia , lookup

Human genetic variation wikipedia , lookup

DNA sequencing wikipedia , lookup

Public health genomics wikipedia , lookup

Human genome wikipedia , lookup

Genomic library wikipedia , lookup

Bisulfite sequencing wikipedia , lookup

Genome evolution wikipedia , lookup

Artificial gene synthesis wikipedia , lookup

Human microbiota wikipedia , lookup

Whole genome sequencing wikipedia , lookup

Human Genome Project wikipedia , lookup

Helitron (biology) wikipedia , lookup

Pathogenomics wikipedia , lookup

Biohistory wikipedia , lookup

RNA-Seq wikipedia , lookup

Genomics wikipedia , lookup

Metagenomics wikipedia , lookup

Transcript
2014 Spring Seminar Speaker Series
Xuegong Zhang, PhD
Professor of Pattern Recognition
and Bioinformatics,
Tsinghua University
Thursday, April 10, 2014
3:30 pm, A115 Crabtree Hall
Alignment-free Machine Learning Analysis of Metagenome Sequencing
Data and a Pilot Study of Microbiome Features on the Tongue
Metagenomes are the mixture of DNAs from all microbial genomes (the microbiome) in samples of environment or
human niches. The next-generation sequencing (NGS) technology has made large-scale study of metagenomes
feasible, which opens a promising new way for understanding our “other self”: the microbiomes that live with us.
Comparing and discriminating metagenome samples is a basic task on analyzing metagenome samples. The
conventional approach based on mapping metagenome sequences to reference genomes and/or genes in databases is
limited by the availability of microbial genomes and gene annotations. An alternative approach is to use sequence
signatures as features to explore the relation among multiple metagenome samples. Typical sequence features are the
relative frequency of different k-mer sequence strings in the metagenome. We conducted a systematic study on the
application of unsupervised and supervised machine learning methods based on sequence features for clustering and
classifying metagenomic samples and illustrated the effectiveness of such reference-free methods in revealing
underlying relationships of the studied samples based on metagenomic sequence features of the microbiomes they
host.
To investigate the relation of microbiomes with human health especially with some characteristics defined in
Traditional Chinese Medicine (TCM), we conducted a pilot study to probe the possible association of microbiome
features on the tongue coating with the Zheng (Syndrome) as described in TCM, using 16S rRNA sequencing and
bioinformatics analysis. Some interesting observations were made, which suggests a new angle to study some basic
TCM abstractions of the human system.