The aim of this study was to analyze the proteome characteristic pattern of unstable angina with blood stasis symptom. Plasma samples were obtained from twelve unstable angina patients and twelve healthy volunteers. To remove the six most abundant proteins, a polyclonal antibody affinity column was used. Then, the two classes of samples were separated by 2D- DIGE. The differentially expressed protein spots were selected and identified with matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) or MS-MS. In the end, using least angle regression algorithm, we studied the proteome characteristic pattern of unstable angina with blood stasis symptom. There are significant difference between unstable angina patients and healthy volunteers. The seventeen proteins made pattern could distinguish unstable angina with qi deficiency and blood stasis syndrome patients from the healthy people and it is probably the proteome characteristic pattern of unstable angina patients with qi deficiency and blood stasis syndrome; The twelve proteins made pattern could distinguish unstable angina with intermingled phlegm and blood stasis syndrome patients from the healthy people and it is probably the proteome characteristic pattern of unstable angina patients with intermingled phlegm and blood stasis syndrome. Using the seventeen proteins made pattern, the unstable angina with qi deficiency and blood stasis syndrome diagnosis accuracy could reach 100 % . Using the twelve proteins made pattern, the unstable angina with intermingled phlegm and blood stasis syndrome diagnosis accuracy also could reach 100%. The least angle regression may be a suitable data mining method for the discovery of illness diagnosis pattern.