Phenotype Classification of Zebrafish Embryos by Supervised Learning

Zebrafish is increasingly used to assess biological properties of chemical substances and thus is becoming a specific tool for toxicological and pharmacological studies. The effects of chemical substances on embryo survival and development are generally evaluated manually through microscopic observation by an expert and documented by several typical photographs. Here, we present a methodology to automatically classify brightfield images of wildtype zebrafish embryos according to their defects by using an image analysis approach based on supervised machine learning. We show that, compared to manual classification, automatic classification results in 90 to 100% agreement with consensus voting of biological experts in nine out of eleven considered defects in 3 days old zebrafish larvae. Automation of the analysis and classification of zebrafish embryo pictures reduces the workload and time required for the biological expert and increases the reproducibility and objectivity of this classification.     

E alpha u and E beta u chain association: where lies the anomaly?

H-2u haplotype mice are unique among all E alpha+ strains because they do not provide in heterozygotes an E alpha chain that interacts with E betak,s,etc. sufficiently well to allow certain E-restricted immune responses. As a first step in understanding this peculiarity, we have sequenced E alpha u and E beta u cDNA and compared the derived amino acid sequences with those of previously analyzed alleles. Although no glaring structural abnormalities were found, we have identified some u-specific residues and suggest which are the most likely to provoke a pairing anomaly.     

Superantigens interact with MHC class II molecules outside of the antigen groove

Superantigens, including the staphylococcal enterotoxins and the minor lymphocyte stimulatory antigens, are highly potent immunostimulatory molecules, capable of activating virtually all T cells that express particular T cell receptor (TCR) variable regions. Superantigen stimulation of T lymphocytes depends on major histocompatibility complex (MHC) class II molecules, so there has been some debate as to whether superantigens interact with the antigen binding "groove" on class II complexes, just like conventional peptide antigens, or whether they bind elsewhere and serve as TCR coligands. We compared the presentation of peptide antigens and superantigens by a panel of mutant-presenting cell lines, each displaying an A kappa alpha chain with a single alanine replacement along the alpha helix proposed to form one face of the groove. The negligible effect of these 30 mutations on superantigen presentation, versus their drastic consequences for peptide presentation, prompts us to conclude that superantigens interact with MHC class II molecules outside the groove.     

Amino acids specifying MHC class preference in TCR V alpha 2 regions.

Some TCR variable regions are preferentially expressed in CD4+ or CD8+ T cells, reflecting a predilection for interacting with MHC class II or class I molecules. The molecular basis for MHC class bias has been studied previously, in particular for V alpha 3 family members, pointing to a dominant role for two amino acid positions in complementary-determining regions (CDRs) 1 and 2. We have evaluated the generality of these findings by examining the MHC class bias of V alpha 2 family members, an attractive system because it shows more variability within the CDR1 and -2, exhibits variation in the framework regions, and includes a member for which the crystal structure has been determined. We find that preferential recognition of MHC class I or II molecules does not always depend on residues at the same positions of CDR1 and -2; rules for one family may be reversed in another. Instead, there are multiple influences exerted by various CDR1/2 positions as well as the CDR3s of both the TCR alpha- and TCR beta-chains.