Organization of Sporulated Oocysts of Eimeria funduli in the Gulf Killifish, Fundulus grandis

Oocysts of Eimeria funduli were studied by transmission electron microscopy in naturally-infected livers of the Gulf killifish, Fundulus grandis. Tissues were cryo-processed because membranous structures in the oocyst appear to hinder routine fixation and embedment. The oocyst wall (about 25 nm thick) was adjacent to the host cell and consisted of an outer membrane that limited the host cell cytoplasm and an inner membrane separated from the outer membrane by a narrow space. In some specimens, dense material was applied to the inner face of the inner membrane. Individual sporocysts were surrounded by a membranous "veil" (about 25 nm thick) that consisted of two unit membranes. Sporopodia, projections of the sporocyst wall, supported the veil. The sporocyst wall (130-150 nm thick) consisted of two layers, a thin electron-lucent outer layer (about 10 nm thick) and a thick electron dense inner layer (about 130 nm thick). Depending on the plane of section, the inner layer had transverse striations with periods of 3 to 4 nm or 12 to 15 nm. A narrow fissure, broadest at the anterior pole of the sporocyst, extended about one-third the length of the sporocyst wall. The posterior pole of the sporocyst was characterized by a bulbous swelling. Although this swelling resembled a Stieda body in light microscopic preparations, ultrastructurally, the swelling was a knoblike thickening in the sporocyst wall and did not plug a gap in this wall

An Extended Kernel for Generalized Multiple-Instance Learning

The multiple-instance learning (MIL) model has been successful in areas such as drug discovery and content-based image-retrieval. Recently, this model was generalized and a corresponding kernel was introduced to learn generalized MIL concepts with a support vector machine. While this kernel enjoyed empirical success, it has limitations in its representation. We extend this kernel by enriching its representation and empirically evaluate our new kernel on data from content-based image retrieval, biological sequence analysis, and drug discovery. We found that our new kernel generalized noticeably better than the old one in content-based image retrieval and biological sequence analysis and was slightly better or even with the old kernel in the other applications, showing that an SVM using this kernel does not overfit despite its richer representation.