Microtubule-associated protein (MAP) tau is a well-established hallmark of a large group of age related neurodegenerative diseases collectively called tauopathies. Under pathological conditions the equilibrium of tau binding to the MTs is perturbed, either by misregulation in the expression levels of specific tau isoforms or by MAPT gene mutations. Preclinical detection of such misregulated tau proteins in cerebrospinal fluid (CSF) is desirable for differential diagnosis and effective prognosis of neurodegeneration. Conventional tau protein detection methods utilize tau isoform-specific antibodies. Such immuno-based protocols, including enzyme-linked immunosorbent assay (ELISA) and Western blots have appropriate sensitivity and specificity, but often show high variability and are time consuming. Here, we established a non-immuno tau protein detection method utilizing microtubule (MT)-kinesin “gliding”assay. All the six tau isoforms expressed in the human brain (0N3R, 1N3R, 2N3R, 0N4R, 1N4R and 2N4R) and five MAPT gene mutants (V248L, G272V, P301L, V337M and R406W) were studied. The landing rate, binding density and gliding velocity of MTs with respect to each tau type were determined and are proposed as tau detection parameters. The detection parameters depicted the type of tau bound to the MTs. Furthermore, MT landing rate and density were found to be superior to gliding velocity in differentiating tau isoforms and mutants. The 3R vs. 4R isoforms, their admixtures, wild vs. mutant 2N4R and specific mutants were differentiated. Our data show that MT-kinesin gliding assay provides a convenient, lab-on-a-chip (LOC) compatible and antibody-free protocol for tau protein analysis.