A transformer architecture for retention time prediction in liquid chromatography mass spectrometry-based proteomics

Accurate retention time (RT) prediction is important for spectral library-based analysis in data-independent acquisition mass spectrometry-based proteomics. The deep learning approach has demonstrated superior performance over traditional machine learning methods for this purpose. The transformer architecture is a recent development in deep learning that delivers state-of-the-art performance in many fields such as natural language processing, computer vision, and biology. We assess the performance of the transformer architecture for RT prediction using datasets from five deep learning models Prosit, DeepDIA, AutoRT, DeepPhospho, and AlphaPeptDeep. The experimental results on holdout datasets and independent datasets exhibit state-of-the-art performance of the transformer architecture. The software and evaluation datasets are publicly available for future development in the field.     

The braincase and palate of the tetrapodomorph sarcopterygian mandageria fairfaxi: morphological variability near the fish–tetrapod transition

The braincase of the Late Devonian tristichopterid sarcopterygian Mandageria fairfaxi , from Canowindra, NSW, Australia, differs radically from the conservative pattern present in other 'osteolepiforms' (stem–group tetrapodomorph fishes) and non–dipnoan sarcopterygian fishes in general. The basioccipital region is short, displaced anteriorly, and either unossified or loosely articulated to the exoccipital, leaving most or all of the notochordal tunnel open ventrally. The exoccipital complex, which is developed into a large saddle that would have rested on top of the notochord, carries large, triangular articular facets on its posterior face and appears to have formed part of a functional neck joint, a synovial articulation between the skull and vertebral column that allows the former to rotate against the latter. Such a joint is characteristic of post–Devonian tetrapods, but unknown in other sarcopterygians. We infer that the ventrally open notochordal tunnel allowed gentle flexion of the cranial notochord during (predominantly vertical) rotational movement at the occiput; this is a mechanically unique solution to the problem of creating a mobile neck. Other unusual features of Mandageria include a posteriorly located lateral commissure, and structures on the entopterygoid and lateral commissure that may have been associated with an elaborate spiracular tract.     

Chick begging as a signal: are nestlings honest?

Begging by dependent avian offspring is known to correlate withhunger level, and parents use this as a signal of brood demandto adjust their chick feeding behavior. While there is informationon how each chick adjusts its begging to its own condition,little is known of how chicks adjust to the state of their nestmates. In two experiments we manipulated the competitive environmentof individual European starling (Sturnus vulgaris) chicks byaltering the state of nest mates while holding the state oftarget chicks constant In the first experiment we placed thetarget chick's nest mates in neighboring nests with brood sizesof two, five, or eight chicks. Following the manipulation wereturned them to their own nests and recorded begging behavioron videotape. In the second experiment we separated a targetchick from its siblings and manipulated feeding level in thelaboratory. The siblings were fed at one of three levels; meanwhile,all the target chicks were fed at the intermediate level. Afterthe manipulation we placed the target chicks with their siblingsand recorded their begging in response to an artificial stimulus.In neither experiment was the begging effort of the unmanipulatedtarget chicks affected by the changes in begging behavior oftheir siblings. This result supports the view that begging isa reliable signal of individual chick state and does not involveresponses to the effort of nest mates.     

Physical mapping of the human T-cell receptor beta gene complex,using yeast artificial chromosomes

Yeast artificial chromosomes (YACs) were used to construct a physical map of the germline human T-cell chain gene complex (TCRB). Variable region genes (BV) for the 25 known subfamilies were used as probes to screen the ICRF AM4x YAC library. Of the five positive YACs identified, one YAC designated B3, 820 kilobase pairs (kbp) in size, scored positive for all 25 TCRBV subfamilies plus the constant region genes (BC) when analyzed by pulse field gel electrophoresis. Restriction enzyme mapping of B3 located TCRBV and TCRBC gene regions to 4 Sfi I fragments of 280 110, 90, and 125 kbp and was in accordance with published data. In addition comparison of hybridization results of Sfi I-restricted B3 and genomic DNA from the parental cell line GM1416B revealed identical banding patterns. The data thus showed YAC B3 encoded a complete and unrearranged TCRB gene locus of some 600–620 kbp. The map was further resolved by locating restriction sites for Sal I and Bss HII on B3, giving more precise localization of the individual TCRBV gene families. Flourescent in situ hybridization of B3 to spreads of human metaphase chromosomes localized B3 to 7q35. However, two additional signals were obtained; one attributable to the TCRBV orphon cluster on 9p21, the second to the long arm of chromosome 2. Polymerase chain reaction amplification of a chromosome 2 somatic cell hybrid, using primers for all 25 TCRBV gene families, revealed that the signal was not attributable to a second orphon cluster. It is suggested that B3 is a chimeric YAC with an intact TCRB locus flanked by chromosome 2 sequences.     

A simple method to quantitatively measure polypeptide JHNHα coupling constants from TOCSY or NOESY spectra

A simple linear relationship between the J coupling constant and the linewidth (_1/2) of in-phase NMR peaks has been identified. This relationship permits the rapid and accurate determination of polypeptide J coupling constants from a simple inspection of amide cross peaks in homonuclear ¹H TOCSY or ¹H NOESY spectra. By using the appropriate set of processing parameters we show that J = 0.5(_1/2) – MW/5000 + 1.8 for TOCSY spectra and J = 0.6(_1/2) – MW/5000 – 0.9 for NOESY spectra, where _1/2 is the half-height linewidth in Hz and MW is the molecular weight of the protein in Da. The simplicity of this relationship, combined with the ease with which _1/2 measurements can be made, means that J coupling constants can now be rapidly determined (up to 100 measurements in less than 30 min) without the need for any complex curve-fitting algorithms. Tests on 11 different polypeptides involving more than 650 separate J measurements have shown that this method yields coupling constants with an rmsd error (relative to X-ray data) of less than 0.9 Hz. Furthermore, the correlation coefficient between the predicted NMR coupling constants and those derived from high-resolution X-ray crystal structures is typically better than 0.89. These simple linear relationships have been found to be valid for peptides as small as 1 kDa to proteins as large as 20 kDa. Despite the method's simplicity, these results are comparable to the accuracy and precision of the best techniques published to date.     

Sensitivity to stress in the bivalve Macoma balthica from the most northern (Arctic) to the most southern (French) populations: low sensitivity in Arctic populations because of genetic adaptations?

The stress sensitivity, determined in copper exposureexperiments and in survival in air tests, and thegenetic structure, measured by means of isoenzymeelectrophoresis, were assessed in populations of theBaltic clam Macoma balthica (L.) from itssouthern to its northern distribution limit, in orderto test the hypotheses that near the distributionlimit the clams would be more stress sensitive andwould have a lower genetic variability. Thepopulations in west and north Europe show a stronggenetic resemblance. The populations in the sub-ArcticWhite Sea are genetically slightly different, and showa low stress sensitivity. The populations in theArctic Pechora Sea are genetically very distant fromthe other populations, and show the lowest stresssensitivity. Near the southern distribution limit, inagreement with the hypotheses, genetic variability islow and stress sensitivity high. On the other hand, incontrast to expectation, near the northerndistribution limit, in the populations of the PechoraSea, the genetic variability was higher, thus notreduced, and the stress sensitivity was low comparedto all other populations. Yet, it remains a questionif such is due to gradual physiologicalacclimatization (and ongoing differential selection)or to genetic adaptation.     

Evolution of the mitochondrial DNA control region in the mbuna (Cichlidae) species flock of lake Malawi, East Africa

Considerable controversy has surrounded the application of mitochondrial DNA data to reconstruction of evolutionary relationships among the endemic cichlids of Lake Malawi. Central to this debate has been the issue of whether lineage sorting is complete, and thus whether these data actually reflect species phylogeny, or simply gene genealogy. Review of all mtDNA control region sequences available for members of one monophyletic subset of this species flock, the Malawi rockfishes, or mbuna, strongly indicates that lineage sorting is incomplete: Character-based analyses of these sequences reconstruct gene, not species, interrelationships. Analysis of the pattern of nucleotide substitutions differentiating these mtDNA alleles suggests that pyrimidine residues undergo transition substitutions more often than do purines. Estimation of the magnitude of derived sequence differentiation in light of the reconstructed gene genealogy suggests that the mbuna may be of considerably more recent vintage than previous molecular characterizations have indicated. Received: 6 April 1996 / Accepted: 3 March 1997