LAB­MaiTE GmbH

Launched in 2021, the Freiburg-based biotech star­tup LAB­MaiTE GmbH is ded­ic­ated to stand­ard­iz­ing the integ­ra­tion of machine learn­ing into fully auto­mated labor­at­ory solu­tions. This aims to har­ness the poten­tial of arti­fi­cial intel­li­gence for bio­lo­gic­al chal­lenges in the future.

A spe­cif­ic focus is on optim­iz­ing fer­ment­at­ive biopro­cesses essen­tial for pro­du­cing anti­bi­ot­ics, enzymes, and oth­er bio­gen­ic value products. Select­ing a suit­able microor­gan­ism and sub­sequent optim­iz­a­tion of the cul­ture con­di­tions for the respect­ive strain are cru­cial to achiev­ing high ah so. Accord­ingly, the eco­nom­ic effi­ciency of the over­all pro­cess also depends on the devel­op­ment of the nutri­ent medi­um. There­fore, devel­op­ing a nutri­ent-appro­pri­ate fer­ment­a­tion medi­um is con­sidered a cent­ral but labor-intens­ive part of biopro­cess development.

The fully auto­mated micro­bi­ore­act­or sys­tem AMLA­MaiTE, developed by LAB­MaiTE, can inde­pend­ently pro­duce nutri­ent media from a range of presen­ted stock solu­tions and iter­at­ively test their suit­ab­il­ity without the need for human inter­ven­tion (e.g., for nutri­ent medi­um pro­duc­tion, inocu­la­tion, or sterilization/​cleaning of the cul­ture vessel).

In addi­tion to the high degree of auto­ma­tion, the over­all sys­tem is also char­ac­ter­ized by using an innov­at­ive, iter­at­ive machine learn­ing approach for nutri­ent medi­um devel­op­ment. The machine learn­ing algorithm has shown advant­ages in yield for the tested bac­teri­al cul­tures com­pared to con­ven­tion­al Design of Exper­i­ments meth­ods for nutri­ent medi­um devel­op­ment. There­fore, cus­tom­ers in indus­tri­al bio­tech­no­logy bene­fit in two ways: firstly, the need for highly qual­i­fied spe­cial­ists for exper­i­ment plan­ning, exe­cu­tion, and ana­lys­is is min­im­ized. Secondly, the prof­it­ab­il­ity of their bio­man­u­fac­tur­ing pro­cess is fur­ther improved by the increase in yield.