Comprehensive Prediction of Molecular Recognition in a Combinatorial Chemical Space Using Machine Learning.

In combinatorial chemical approaches, optimizing the composition and arrangement of building blocks toward a particular function has been done using a number of methods, including high throughput molecular screening, molecular evolution, and computational prescreening. Here, a different approach is considered that uses sparse measurements of library molecules as the input to a machine learning algorithm which generates a comprehensive, quantitative relationship between covalent molecular structure and function that can then be used to predict the function of any molecule in the possible combinatorial space. To test the feasibility of the approach, a defined combinatorial chemical space consisting of ∼1012 possible linear combinations of 16 different amino acids was used. The binding of a very sparse, but nearly random, sampling of this amino acid sequence space to 9 different protein targets is measured and used to generate a general relationship between peptide sequence and binding for each target. Surprisingly, measuring as little as a few hundred to a few thousand of the ∼1012 possible molecules provides sufficient training to be highly predictive of the binding of the remaining molecules in the combinatorial space. Furthermore, measuring only amino acid sequences that bind weakly to a target allows the accurate prediction of which sequences will bind 10-100 times more strongly. Thus, the molecular recognition information contained in a tiny fraction of molecules in this combinatorial space is sufficient to characterize any set of molecules randomly selected from the entire space, a fact that potentially has significant implications for the design of new chemical function using combinatorial chemical libraries.

An ImmunoSignature test distinguishes Trypanosoma cruzi, hepatitis B, hepatitis C and West Nile virus seropositivity among asymptomatic blood donors.

BACKGROUND: The complexity of the eukaryotic parasite Trypanosoma (T.) cruzi manifests in its highly dynamic genome, multi-host life cycle, progressive morphologies and immune-evasion mechanisms. Accurate determination of infection or Chagas' disease activity and prognosis continues to challenge researchers. We hypothesized that a diagnostic platform with higher ligand complexity than previously employed may hold value. METHODOLOGY: We applied the ImmunoSignature Technology (IST) for the detection of T. cruzi-specific antibodies among healthy blood donors. IST is based on capturing the information in an individual's antibody repertoire by exposing their peripheral blood to a library of >100,000 position-addressable, chemically-diverse peptides. PRINCIPAL FINDINGS: Initially, samples from two Chagas cohorts declared positive or negative by bank testing were studied. With the first cohort, library-peptides displaying differential binding signals between T. cruzi sero-states were used to train an algorithm. A classifier was fixed and tested against the training-independent second cohort to determine assay performance. Next, samples from a mixed cohort of donors declared positive for Chagas, hepatitis B, hepatitis C or West Nile virus were assayed on the same library. Signals were used to train a single algorithm that distinguished all four disease states. As a binary test, the accuracy of predicting T. cruzi seropositivity by IST was similar, perhaps modestly reduced, relative to conventional ELISAs. However, the results indicate that information beyond determination of seropositivity may have been captured. These include the identification of cohort subclasses, the simultaneous detection and discerning of other diseases, and the discovery of putative new antigens. CONCLUSIONS & SIGNIFICANCE: The central outcome of this study established IST as a reliable approach for specific determination of T. cruzi seropositivity versus disease-free individuals or those with other diseases. Its potential contribution for monitoring and controlling Chagas lies in IST's delivery of higher resolution immune-state readouts than obtained with currently-used technologies. Despite the complexity of the ligand presentation and large quantitative readouts, performing an IST test is simple, scalable and reproducible.

Immunosignaturing: a critical review.

Health is a complex interaction between metabolism, physiology, and immunity. Although it is difficult to define quantitatively, the activity of the humoral immune system provides a reasonable proxy for changes in health. Immunosignaturing is a microarray-based technology that quantitates the dynamics of circulating antibodies. Recent advancements in the field warrant a review of the technology. Here, we provide an introduction to the technique, evaluate the current progress, contrast similar technologies, and suggest applications that immunosignaturing could facilitate.

Analysis of peptide mimotopes of Burkholderia pseudomallei exopolysaccharide.

Previously two capsule-specific monoclonal antibodies (4VA5 and 3VIE5) were identified as protective against Burkholderia pseudomallei in passive transfer experiments. Panning these antibodies against evolutionary phage libraries identified reactive peptides capable of inhibiting its parent monoclonal from binding to B. pseudomallei. Mice immunized with peptide conjugated to thyroglobulin developed serum antibodies capable of recognizing the immunizing peptide of which a subset recognized exopolysaccharide in the context of whole B. pseudomallei cells. These serum antibodies recognized protease treated B. pseudomallei but not B. thailandensis suggesting that these peptides are mimotopes of the B. pseudomallei capsular exopolysaccharide. In a murine model of acute melioidosis, immunization with the mimotope of the 4VA5 binding site extended the mean time to death to 8.00 days over the 2.18 days afforded by immunization with thyroglobulin alone. This mimotope may be of use in developing an antibody response against B. pseudomallei exopolysaccharide.