Community-acquired pneumonia (CAP) is a heterogeneous disease causing great morbidity, mortality and health care burden globally. Typing methods for discriminating different clinical conditions of the same disease are essential to a better management of CAP. Traditional typing systems based separately on clinical manifestations (such as PSI and CURB-65), pathogens(bacterial types, virulence, drug resistance, etc) or host immune state (immunocompetent, immunocompromised or immunodeficiency). Thus, they are barely able to represent the real disease status nor to precisely predict the mortality.

As the development of multi-omic technologies, the relatedness of different phenotypes at a molecular level have revolutionized our ability to differentiate among patients. Our study is aimed at establishing a novel molecular typing method of CAP. Multi-omic (including genomics, transcriptomes, and metabolisms) data obtained from enrolled CAP patients and isolated pathogens would be integrated analyzed and interpreted. Tthe investigators believe that an appropriate molecular typing method would lead to revolutionary changes in current arrangements of CAP.

Inclusion Criteria:

• adult (aged > 16 years)
• diagnosed as community-acquired pneumonia

Exclusion Criteria:

• being immunocompromised, including history of glucocorticoid taken for more than 1 month, history of immunosuppressive therapy, history of human immunodeficiency virus (HIV) infection, solid tumor or hematological malignancy
• history of long-term nursing home stays
• history of recently hospitalized (<90 days)

all adult patients (aged > 16 years) admit to the 4 hospitals between March 2017 and March 2018 with CAP