The early stage of viral infection is often followed by an important increase of viral load and is generally considered to be the most at risk for pathogen transmission. Most methods quantifying the relative importance of the different stages of infection were developed for studies aimed at measuring HIV transmission in Humans. However, they cannot be transposed to animal populations in which less information is available. Here we propose a general method to quantify the importance of the early and late stages of the infection on micro-organism transmission from field studies. The method is based on a state space dynamical model parameterized using Bayesian inference. It is illustrated by a 28 years dataset in mandrills infected by Simian Immunodeficiency Virus type-1 (SIV-1) and the Simian T-Cell Lymphotropic Virus type-1 (STLV-1). For both viruses we show that transmission is predominant during the early stage of the infection (transmission ratio for SIV-1: 1.16 [0.0009; 18.15] and 9.92 [0.03; 83.8] for STLV-1). However, in terms of basic reproductive number (R0 ), which quantifies the weight of both stages in the spread of the virus, the results suggest that the epidemics of SIV-1 and STLV-1 are mainly driven by late transmissions in this population.