TITLE: A collection of genetically engineered Populus trees reveals wood biomass traits that predict glucose yield from enzymatic hydrolysis Short title: Tools to predict glucose yield from Populus wood

Wood represents a promising source of lignocellulosic biomass for the production of bio-based renewables, especially biofuels. However, woody feedstocks must be improved to become competitive against petroleum. We created a collection of Populus trees consisting of 40 genetically engineered lines to modify and to better understand wood biomass properties. A total of 65 traits were measured in these trees and in the corresponding wild-type clone, including growth parameters, wood anatomical and structural properties, cell wall composition and analytical saccharification. The relationships between saccharification of glucose and biomass traits were investigated using multivariate data analysis methods and mathematical modeling. To circumvent potential trade-offs between biomass production and saccharification potential, we also estimated the “total-wood glucose yield” (TWG) expected after pretreatment and 72h of enzymatic hydrolysis from whole trees. A mathematical model estimated TWG from a subset of 22 wood biomass traits with good predictivity (Q = 0.8), while saccharification of glucose could be predicted from seven biomass traits (Q = 0.49). Among the seven diagnostic traits for saccharification, four also affected biomass production, such as the ratio of Sto G-lignin which was beneficial for saccharification but detrimental for growth. The contents of various matrix polysaccharides appeared important for predicting both saccharification and TWG, including low abundance monosaccharides. In particular, fucose and mannose contents negatively correlated with TWG, apparently by negatively associating with biomass production. Both biomass production and saccharification, and hence TWG, negatively correlated with arabinose and rhamnose contents, suggesting that these low abundance monosaccharides represent markers/targets for improving feedstocks.